Merge remote-tracking branch 'dan/master' into deeper-conformer

.flake8

@@ -4,14 +4,11 @@ statistics=true
 max-line-length = 80
 per-file-ignores =
     # line too long
-    egs/librispeech/ASR/*/conformer.py: E501,
+    icefall/diagnostics.py: E501
-    egs/aishell/ASR/*/conformer.py: E501,
+    egs/*/ASR/*/conformer.py: E501,
-    egs/tedlium3/ASR/*/conformer.py: E501,
+    egs/*/ASR/pruned_transducer_stateless*/*.py: E501,
-    egs/gigaspeech/ASR/*/conformer.py: E501,
+    egs/*/ASR/*/optim.py: E501,
-    egs/librispeech/ASR/pruned_transducer_stateless2/*.py: E501,
+    egs/*/ASR/*/scaling.py: E501,
-    egs/librispeech/ASR/pruned_transducer_stateless4/*.py: E501,
-    egs/librispeech/ASR/*/optim.py: E501,
-    egs/librispeech/ASR/*/scaling.py: E501,
 
     # invalid escape sequence (cause by tex formular), W605
     icefall/utils.py: E501, W605

.github/scripts/download-gigaspeech-dev-test-dataset.sh

@@ -0,0 +1,15 @@
+#!/usr/bin/env bash
+
+# This script downloads the pre-computed fbank features for
+# dev and test datasets of GigaSpeech.
+#
+# You will find directories `~/tmp/giga-dev-dataset-fbank` after running
+# this script.
+
+mkdir -p ~/tmp
+cd ~/tmp
+
+git lfs install
+git clone https://huggingface.co/csukuangfj/giga-dev-dataset-fbank
+
+ls -lh giga-dev-dataset-fbank/data/fbank

.github/scripts/run-gigaspeech-pruned-transducer-stateless2-2022-05-12.sh

@@ -0,0 +1,49 @@
+#!/usr/bin/env bash
+
+log() {
+  # This function is from espnet
+  local fname=${BASH_SOURCE[1]##*/}
+  echo -e "$(date '+%Y-%m-%d %H:%M:%S') (${fname}:${BASH_LINENO[0]}:${FUNCNAME[1]}) $*"
+}
+
+cd egs/gigaspeech/ASR
+
+repo_url=https://huggingface.co/wgb14/icefall-asr-gigaspeech-pruned-transducer-stateless2
+
+log "Downloading pre-trained model from $repo_url"
+git lfs install
+git clone $repo_url
+repo=$(basename $repo_url)
+
+echo "GITHUB_EVENT_NAME: ${GITHUB_EVENT_NAME}"
+echo "GITHUB_EVENT_LABEL_NAME: ${GITHUB_EVENT_LABEL_NAME}"
+if [[ x"${GITHUB_EVENT_NAME}" == x"schedule" || x"${GITHUB_EVENT_LABEL_NAME}" == x"run-decode"  ]]; then
+  mkdir -p pruned_transducer_stateless2/exp
+  ln -s $PWD/$repo/exp/pretrained-iter-3488000-avg-20.pt pruned_transducer_stateless2/exp/epoch-999.pt
+  ln -s $PWD/$repo/data/lang_bpe_500 data/
+
+  ls -lh data
+  ls -lh data/lang_bpe_500
+  ls -lh data/fbank
+  ls -lh pruned_transducer_stateless2/exp
+
+  log "Decoding dev and test"
+
+  # use a small value for decoding with CPU
+  max_duration=100
+
+  # Test only greedy_search to reduce CI running time
+  # for method in greedy_search fast_beam_search modified_beam_search; do
+  for method in greedy_search; do
+    log "Decoding with $method"
+
+    ./pruned_transducer_stateless2/decode.py \
+      --decoding-method $method \
+      --epoch 999 \
+      --avg 1 \
+      --max-duration $max_duration \
+      --exp-dir pruned_transducer_stateless2/exp
+  done
+
+  rm pruned_transducer_stateless2/exp/*.pt
+fi

.github/scripts/run-librispeech-pruned-transducer-stateless3-2022-05-13.sh

@@ -0,0 +1,80 @@
+#!/usr/bin/env bash
+
+log() {
+  # This function is from espnet
+  local fname=${BASH_SOURCE[1]##*/}
+  echo -e "$(date '+%Y-%m-%d %H:%M:%S') (${fname}:${BASH_LINENO[0]}:${FUNCNAME[1]}) $*"
+}
+
+cd egs/librispeech/ASR
+
+repo_url=https://huggingface.co/csukuangfj/icefall-asr-librispeech-pruned-transducer-stateless3-2022-05-13
+
+log "Downloading pre-trained model from $repo_url"
+git lfs install
+git clone $repo_url
+repo=$(basename $repo_url)
+
+log "Display test files"
+tree $repo/
+soxi $repo/test_wavs/*.wav
+ls -lh $repo/test_wavs/*.wav
+
+pushd $repo/exp
+ln -s pretrained-iter-1224000-avg-14.pt pretrained.pt
+popd
+
+for sym in 1 2 3; do
+  log "Greedy search with --max-sym-per-frame $sym"
+
+  ./pruned_transducer_stateless3/pretrained.py \
+    --method greedy_search \
+    --max-sym-per-frame $sym \
+    --checkpoint $repo/exp/pretrained.pt \
+    --bpe-model $repo/data/lang_bpe_500/bpe.model \
+    $repo/test_wavs/1089-134686-0001.wav \
+    $repo/test_wavs/1221-135766-0001.wav \
+    $repo/test_wavs/1221-135766-0002.wav
+done
+
+for method in modified_beam_search beam_search fast_beam_search; do
+  log "$method"
+
+  ./pruned_transducer_stateless3/pretrained.py \
+    --method $method \
+    --beam-size 4 \
+    --checkpoint $repo/exp/pretrained.pt \
+    --bpe-model $repo/data/lang_bpe_500/bpe.model \
+    $repo/test_wavs/1089-134686-0001.wav \
+    $repo/test_wavs/1221-135766-0001.wav \
+    $repo/test_wavs/1221-135766-0002.wav
+done
+
+echo "GITHUB_EVENT_NAME: ${GITHUB_EVENT_NAME}"
+echo "GITHUB_EVENT_LABEL_NAME: ${GITHUB_EVENT_LABEL_NAME}"
+if [[ x"${GITHUB_EVENT_NAME}" == x"schedule" || x"${GITHUB_EVENT_LABEL_NAME}" == x"run-decode"  ]]; then
+  mkdir -p pruned_transducer_stateless3/exp
+  ln -s $PWD/$repo/exp/pretrained.pt pruned_transducer_stateless3/exp/epoch-999.pt
+  ln -s $PWD/$repo/data/lang_bpe_500 data/
+
+  ls -lh data
+  ls -lh pruned_transducer_stateless3/exp
+
+  log "Decoding test-clean and test-other"
+
+  # use a small value for decoding with CPU
+  max_duration=100
+
+  for method in greedy_search fast_beam_search modified_beam_search; do
+    log "Decoding with $method"
+
+    ./pruned_transducer_stateless3/decode.py \
+      --decoding-method $method \
+      --epoch 999 \
+      --avg 1 \
+      --max-duration $max_duration \
+      --exp-dir pruned_transducer_stateless3/exp
+  done
+
+  rm pruned_transducer_stateless3/exp/*.pt
+fi

.github/workflows/run-gigaspeech-2022-05-13.yml

@@ -0,0 +1,120 @@
+# Copyright      2021  Fangjun Kuang (csukuangfj@gmail.com)
+
+# See ../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+name: run-gigaspeech-2022-05-13
+# stateless transducer + k2 pruned rnnt-loss + reworked conformer
+
+on:
+  push:
+    branches:
+      - master
+  pull_request:
+    types: [labeled]
+
+  schedule:
+    # minute (0-59)
+    # hour (0-23)
+    # day of the month (1-31)
+    # month (1-12)
+    # day of the week (0-6)
+    # nightly build at 15:50 UTC time every day
+    - cron: "50 15 * * *"
+
+jobs:
+  run_gigaspeech_2022_05_13:
+    if: github.event.label.name == 'ready' || github.event.label.name == 'run-decode' || github.event_name == 'push' || github.event_name == 'schedule'
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os: [ubuntu-18.04]
+        python-version: [3.7, 3.8, 3.9]
+
+      fail-fast: false
+
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: 0
+
+      - name: Setup Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v2
+        with:
+          python-version: ${{ matrix.python-version }}
+          cache: 'pip'
+          cache-dependency-path: '**/requirements-ci.txt'
+
+      - name: Install Python dependencies
+        run: |
+          grep -v '^#' ./requirements-ci.txt  | xargs -n 1 -L 1 pip install
+
+      - name: Cache kaldifeat
+        id: my-cache
+        uses: actions/cache@v2
+        with:
+          path: |
+            ~/tmp/kaldifeat
+          key: cache-tmp-${{ matrix.python-version }}
+
+      - name: Install kaldifeat
+        if: steps.my-cache.outputs.cache-hit != 'true'
+        shell: bash
+        run: |
+          .github/scripts/install-kaldifeat.sh
+
+      - name: Download GigaSpeech dev/test dataset
+        shell: bash
+        run: |
+          sudo apt-get install -y -q git-lfs
+
+          .github/scripts/download-gigaspeech-dev-test-dataset.sh
+
+      - name: Inference with pre-trained model
+        shell: bash
+        env:
+          GITHUB_EVENT_NAME: ${{ github.event_name }}
+          GITHUB_EVENT_LABEL_NAME: ${{ github.event.label.name }}
+        run: |
+          ln -s ~/tmp/giga-dev-dataset-fbank/data egs/gigaspeech/ASR/
+
+          ls -lh egs/gigaspeech/ASR/data/fbank
+
+          export PYTHONPATH=$PWD:$PYTHONPATH
+          export PYTHONPATH=~/tmp/kaldifeat/kaldifeat/python:$PYTHONPATH
+          export PYTHONPATH=~/tmp/kaldifeat/build/lib:$PYTHONPATH
+
+          .github/scripts/run-gigaspeech-pruned-transducer-stateless2-2022-05-12.sh
+
+      - name: Display decoding results for gigaspeech pruned_transducer_stateless2
+        if: github.event_name == 'schedule' || github.event.label.name == 'run-decode'
+        shell: bash
+        run: |
+          cd egs/gigaspeech/ASR/
+          tree ./pruned_transducer_stateless2/exp
+
+          sudo apt-get -qq install tree
+
+          cd pruned_transducer_stateless2
+          echo "results for pruned_transducer_stateless2"
+          echo "===greedy search==="
+          find exp/greedy_search -name "log-*" -exec grep -n --color "best for dev" {} + | sort -n -k2
+          find exp/greedy_search -name "log-*" -exec grep -n --color "best for test" {} + | sort -n -k2
+
+      - name: Upload decoding results for gigaspeech pruned_transducer_stateless2
+        uses: actions/upload-artifact@v2
+        if: github.event_name == 'schedule' || github.event.label.name == 'run-decode'
+        with:
+          name: torch-${{ matrix.torch }}-python-${{ matrix.python-version }}-ubuntu-18.04-cpu-gigaspeech-pruned_transducer_stateless2-2022-05-12
+          path: egs/gigaspeech/ASR/pruned_transducer_stateless2/exp/

.github/workflows/run-librispeech-2022-04-29.yml

@@ -142,8 +142,8 @@ jobs:
           find fast_beam_search -name "log-*" -exec grep -n --color "best for test-other" {} + | sort -n -k2
 
           echo "===modified beam search==="
-          find exp/modified_beam_search -name "log-*" -exec grep -n --color "best for test-clean" {} + | sort -n -k2
+          find modified_beam_search -name "log-*" -exec grep -n --color "best for test-clean" {} + | sort -n -k2
-          find exp/modified_beam_search -name "log-*" -exec grep -n --color "best for test-other" {} + | sort -n -k2
+          find modified_beam_search -name "log-*" -exec grep -n --color "best for test-other" {} + | sort -n -k2
 
       - name: Display decoding results for pruned_transducer_stateless3
         if: github.event_name == 'schedule' || github.event.label.name == 'run-decode'
@@ -161,8 +161,8 @@ jobs:
           find fast_beam_search -name "log-*" -exec grep -n --color "best for test-other" {} + | sort -n -k2
 
           echo "===modified beam search==="
-          find exp/modified_beam_search -name "log-*" -exec grep -n --color "best for test-clean" {} + | sort -n -k2
+          find modified_beam_search -name "log-*" -exec grep -n --color "best for test-clean" {} + | sort -n -k2
-          find exp/modified_beam_search -name "log-*" -exec grep -n --color "best for test-other" {} + | sort -n -k2
+          find modified_beam_search -name "log-*" -exec grep -n --color "best for test-other" {} + | sort -n -k2
 
       - name: Upload decoding results for pruned_transducer_stateless2
         uses: actions/upload-artifact@v2

.github/workflows/run-librispeech-pruned-transducer-stateless3-2022-05-13.yml

@@ -0,0 +1,151 @@
+# Copyright      2021  Fangjun Kuang (csukuangfj@gmail.com)
+
+# See ../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+name: run-librispeech-pruned-transducer-stateless3-2022-05-13
+# stateless pruned transducer (reworked model) + giga speech
+
+on:
+  push:
+    branches:
+      - master
+  pull_request:
+    types: [labeled]
+
+  schedule:
+    # minute (0-59)
+    # hour (0-23)
+    # day of the month (1-31)
+    # month (1-12)
+    # day of the week (0-6)
+    # nightly build at 15:50 UTC time every day
+    - cron: "50 15 * * *"
+
+jobs:
+  run_librispeech_pruned_transducer_stateless3_2022_05_13:
+    if: github.event.label.name == 'ready' || github.event.label.name == 'run-decode' || github.event_name == 'push' || github.event_name == 'schedule'
+    runs-on: ${{ matrix.os }}
+    strategy:
+      matrix:
+        os: [ubuntu-18.04]
+        python-version: [3.7, 3.8, 3.9]
+
+      fail-fast: false
+
+    steps:
+      - uses: actions/checkout@v2
+        with:
+          fetch-depth: 0
+
+      - name: Setup Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v2
+        with:
+          python-version: ${{ matrix.python-version }}
+          cache: 'pip'
+          cache-dependency-path: '**/requirements-ci.txt'
+
+      - name: Install Python dependencies
+        run: |
+          grep -v '^#' ./requirements-ci.txt  | xargs -n 1 -L 1 pip install
+
+      - name: Cache kaldifeat
+        id: my-cache
+        uses: actions/cache@v2
+        with:
+          path: |
+            ~/tmp/kaldifeat
+          key: cache-tmp-${{ matrix.python-version }}
+
+      - name: Install kaldifeat
+        if: steps.my-cache.outputs.cache-hit != 'true'
+        shell: bash
+        run: |
+          .github/scripts/install-kaldifeat.sh
+
+      - name: Cache LibriSpeech test-clean and test-other datasets
+        id: libri-test-clean-and-test-other-data
+        uses: actions/cache@v2
+        with:
+          path: |
+            ~/tmp/download
+          key: cache-libri-test-clean-and-test-other
+
+      - name: Download LibriSpeech test-clean and test-other
+        if: steps.libri-test-clean-and-test-other-data.outputs.cache-hit != 'true'
+        shell: bash
+        run: |
+          .github/scripts/download-librispeech-test-clean-and-test-other-dataset.sh
+
+      - name: Prepare manifests for LibriSpeech test-clean and test-other
+        shell: bash
+        run: |
+          .github/scripts/prepare-librispeech-test-clean-and-test-other-manifests.sh
+
+      - name: Cache LibriSpeech test-clean and test-other fbank features
+        id: libri-test-clean-and-test-other-fbank
+        uses: actions/cache@v2
+        with:
+          path: |
+            ~/tmp/fbank-libri
+          key: cache-libri-fbank-test-clean-and-test-other
+
+      - name: Compute fbank for LibriSpeech test-clean and test-other
+        if: steps.libri-test-clean-and-test-other-fbank.outputs.cache-hit != 'true'
+        shell: bash
+        run: |
+          .github/scripts/compute-fbank-librispeech-test-clean-and-test-other.sh
+
+      - name: Inference with pre-trained model
+        shell: bash
+        env:
+          GITHUB_EVENT_NAME: ${{ github.event_name }}
+          GITHUB_EVENT_LABEL_NAME: ${{ github.event.label.name }}
+        run: |
+          mkdir -p egs/librispeech/ASR/data
+          ln -sfv ~/tmp/fbank-libri egs/librispeech/ASR/data/fbank
+          ls -lh egs/librispeech/ASR/data/*
+
+          sudo apt-get -qq install git-lfs tree sox
+          export PYTHONPATH=$PWD:$PYTHONPATH
+          export PYTHONPATH=~/tmp/kaldifeat/kaldifeat/python:$PYTHONPATH
+          export PYTHONPATH=~/tmp/kaldifeat/build/lib:$PYTHONPATH
+
+          .github/scripts/run-librispeech-pruned-transducer-stateless3-2022-05-13.sh
+
+      - name: Display decoding results for pruned_transducer_stateless3
+        if: github.event_name == 'schedule' || github.event.label.name == 'run-decode'
+        shell: bash
+        run: |
+          cd egs/librispeech/ASR
+          tree pruned_transducer_stateless3/exp
+          cd pruned_transducer_stateless3/exp
+          echo "===greedy search==="
+          find greedy_search -name "log-*" -exec grep -n --color "best for test-clean" {} + | sort -n -k2
+          find greedy_search -name "log-*" -exec grep -n --color "best for test-other" {} + | sort -n -k2
+
+          echo "===fast_beam_search==="
+          find fast_beam_search -name "log-*" -exec grep -n --color "best for test-clean" {} + | sort -n -k2
+          find fast_beam_search -name "log-*" -exec grep -n --color "best for test-other" {} + | sort -n -k2
+
+          echo "===modified beam search==="
+          find modified_beam_search -name "log-*" -exec grep -n --color "best for test-clean" {} + | sort -n -k2
+          find modified_beam_search -name "log-*" -exec grep -n --color "best for test-other" {} + | sort -n -k2
+
+      - name: Upload decoding results for pruned_transducer_stateless3
+        uses: actions/upload-artifact@v2
+        if: github.event_name == 'schedule' || github.event.label.name == 'run-decode'
+        with:
+          name: torch-${{ matrix.torch }}-python-${{ matrix.python-version }}-ubuntu-18.04-cpu-pruned_transducer_stateless3-2022-04-29
+          path: egs/librispeech/ASR/pruned_transducer_stateless3/exp/

.github/workflows/test.yml

@@ -103,11 +103,26 @@ jobs:
           cd egs/librispeech/ASR/conformer_ctc
           pytest -v -s
 
+          cd ../pruned_transducer_stateless
+          pytest -v -s
+
+          cd ../pruned_transducer_stateless2
+          pytest -v -s
+
+          cd ../pruned_transducer_stateless3
+          pytest -v -s
+
+          cd ../pruned_transducer_stateless4
+          pytest -v -s
+
+          cd ../transducer_stateless
+          pytest -v -s
+
           if [[ ${{ matrix.torchaudio }} == "0.10.0" ]]; then
             cd ../transducer
             pytest -v -s
 
-            cd ../transducer_stateless
+            cd ../transducer_stateless2
             pytest -v -s
 
             cd ../transducer_lstm
@@ -128,11 +143,26 @@ jobs:
           cd egs/librispeech/ASR/conformer_ctc
           pytest -v -s
 
+          cd ../pruned_transducer_stateless
+          pytest -v -s
+
+          cd ../pruned_transducer_stateless2
+          pytest -v -s
+
+          cd ../pruned_transducer_stateless3
+          pytest -v -s
+
+          cd ../pruned_transducer_stateless4
+          pytest -v -s
+
+          cd ../transducer_stateless
+          pytest -v -s
+
           if [[ ${{ matrix.torchaudio }} == "0.10.0" ]]; then
             cd ../transducer
             pytest -v -s
 
-            cd ../transducer_stateless
+            cd ../transducer_stateless2
             pytest -v -s
 
             cd ../transducer_lstm

README.md

@@ -12,13 +12,14 @@ for installation.
 Please refer to <https://icefall.readthedocs.io/en/latest/recipes/index.html>
 for more information.
 
-We provide four recipes at present:
+We provide 6 recipes at present:
 
   - [yesno][yesno]
   - [LibriSpeech][librispeech]
   - [Aishell][aishell]
   - [TIMIT][timit]
   - [TED-LIUM3][tedlium3]
+  - [GigaSpeech][gigaspeech]
 
 ### yesno
 
@@ -106,7 +107,7 @@ We provide a Colab notebook to run a pre-trained transducer conformer + stateles
 
 |     | test-clean | test-other |
 |-----|------------|------------|
-| WER | 2.19       | 4.97       |
+| WER | 2.00       | 4.63       |
 
 
 ### Aishell
@@ -197,6 +198,26 @@ The best WER using modified beam search with beam size 4 is:
 
 We provide a Colab notebook to run a pre-trained Pruned Transducer Stateless model: [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1je_1zGrOkGVVd4WLzgkXRHxl-I27yWtz?usp=sharing)
 
+### GigaSpeech
+
+We provide two models for this recipe: [Conformer CTC model][GigaSpeech_conformer_ctc]
+and [Pruned stateless RNN-T: Conformer encoder + Embedding decoder + k2 pruned RNN-T loss][GigaSpeech_pruned_transducer_stateless2].
+
+#### Conformer CTC
+
+|     |  Dev  | Test  |
+|-----|-------|-------|
+| WER | 10.47 | 10.58 |
+
+#### Pruned stateless RNN-T: Conformer encoder + Embedding decoder + k2 pruned RNN-T loss
+
+|                      |  Dev  | Test  |
+|----------------------|-------|-------|
+|    greedy search     | 10.51 | 10.73 |
+|   fast beam search   | 10.50 | 10.69 |
+| modified beam search | 10.40 | 10.51 |
+
+
 ## Deployment with C++
 
 Once you have trained a model in icefall, you may want to deploy it with C++,
@@ -220,9 +241,12 @@ Please see: [![Open In Colab](https://colab.research.google.com/assets/colab-bad
 [TIMIT_tdnn_ligru_ctc]: egs/timit/ASR/tdnn_ligru_ctc
 [TED-LIUM3_transducer_stateless]: egs/tedlium3/ASR/transducer_stateless
 [TED-LIUM3_pruned_transducer_stateless]: egs/tedlium3/ASR/pruned_transducer_stateless
+[GigaSpeech_conformer_ctc]: egs/gigaspeech/ASR/conformer_ctc
+[GigaSpeech_pruned_transducer_stateless2]: egs/gigaspeech/ASR/pruned_transducer_stateless2
 [yesno]: egs/yesno/ASR
 [librispeech]: egs/librispeech/ASR
 [aishell]: egs/aishell/ASR
 [timit]: egs/timit/ASR
 [tedlium3]: egs/tedlium3/ASR
+[gigaspeech]: egs/gigaspeech/ASR
 [k2]: https://github.com/k2-fsa/k2

egs/gigaspeech/ASR/README.md

@@ -13,8 +13,9 @@ ln -sfv /path/to/GigaSpeech download/GigaSpeech
 ```
 
 ## Performance Record
-|     |  Dev  | Test  |
+|                                |  Dev  | Test  |
-|-----|-------|-------|
+|--------------------------------|-------|-------|
-| WER | 10.47 | 10.58 |
+|         `conformer_ctc`        | 10.47 | 10.58 |
+| `pruned_transducer_stateless2` | 10.40 | 10.51 |
 
 See [RESULTS](/egs/gigaspeech/ASR/RESULTS.md) for details.

egs/gigaspeech/ASR/RESULTS.md

@@ -1,4 +1,77 @@
 ## Results
+### GigaSpeech BPE training results (Pruned Transducer 2)
+
+#### 2022-05-12
+
+#### Conformer encoder + embedding decoder
+
+Conformer encoder + non-recurrent decoder. The encoder is a
+reworked version of the conformer encoder, with many changes. The
+decoder contains only an embedding layer, a Conv1d (with kernel
+size 2) and a linear layer (to transform tensor dim). k2 pruned
+RNN-T loss is used.
+
+The best WER, as of 2022-05-12, for the gigaspeech is below
+
+Results are:
+
+|                      |  Dev  | Test  |
+|----------------------|-------|-------|
+|    greedy search     | 10.51 | 10.73 |
+|   fast beam search   | 10.50 | 10.69 |
+| modified beam search | 10.40 | 10.51 |
+
+To reproduce the above result, use the following commands for training:
+
+```bash
+cd egs/gigaspeech/ASR
+./prepare.sh
+export CUDA_VISIBLE_DEVICES="0,1,2,3,4,5,6,7"
+./pruned_transducer_stateless2/train.py \
+  --max-duration 120 \
+  --num-workers 1 \
+  --world-size 8 \
+  --exp-dir pruned_transducer_stateless2/exp \
+  --bpe-model data/lang_bpe_500/bpe.model \
+  --use-fp16 True
+```
+
+and the following commands for decoding:
+
+```bash
+# greedy search
+./pruned_transducer_stateless2/decode.py \
+  --iter 3488000 \
+  --avg 20 \
+  --decoding-method greedy_search \
+  --exp-dir pruned_transducer_stateless2/exp \
+  --bpe-model data/lang_bpe_500/bpe.model \
+  --max-duration 600
+
+# fast beam search
+./pruned_transducer_stateless2/decode.py \
+  --iter 3488000 \
+  --avg 20 \
+  --decoding-method fast_beam_search \
+  --exp-dir pruned_transducer_stateless2/exp \
+  --bpe-model data/lang_bpe_500/bpe.model \
+  --max-duration 600
+
+# modified beam search
+./pruned_transducer_stateless2/decode.py \
+  --iter 3488000 \
+  --avg 15 \
+  --decoding-method modified_beam_search \
+  --exp-dir pruned_transducer_stateless2/exp \
+  --bpe-model data/lang_bpe_500/bpe.model \
+  --max-duration 600
+```
+
+Pretrained model is available at
+<https://huggingface.co/wgb14/icefall-asr-gigaspeech-pruned-transducer-stateless2>
+
+The tensorboard log for training is available at
+<https://tensorboard.dev/experiment/zmmM0MLASnG1N2RmJ4MZBw/>
 
 ### GigaSpeech BPE training results (Conformer-CTC)
 
@@ -20,7 +93,7 @@ Scale values used in n-gram LM rescoring and attention rescoring for the best WE
 
 To reproduce the above result, use the following commands for training:
 
-```
+```bash
 cd egs/gigaspeech/ASR
 ./prepare.sh
 export CUDA_VISIBLE_DEVICES="0,1,2,3,4,5,6,7"
@@ -34,7 +107,7 @@ export CUDA_VISIBLE_DEVICES="0,1,2,3,4,5,6,7"
 
 and the following command for decoding:
 
-```
+```bash
 ./conformer_ctc/decode.py \
   --epoch 18 \
   --avg 6 \
@@ -59,7 +132,7 @@ Scale values used in n-gram LM rescoring and attention rescoring for the best WE
 
 To reproduce the above result, use the training commands above, and the following command for decoding:
 
-```
+```bash
 ./conformer_ctc/decode.py \
   --epoch 18 \
   --avg 6 \

egs/gigaspeech/ASR/pruned_transducer_stateless2/asr_datamodule.py

@@ -0,0 +1,416 @@
+# Copyright      2021  Piotr Żelasko
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+import argparse
+import inspect
+import logging
+from functools import lru_cache
+from pathlib import Path
+from typing import Any, Dict, Optional
+
+import torch
+from lhotse import CutSet, Fbank, FbankConfig, load_manifest
+from lhotse.dataset import (
+    BucketingSampler,
+    CutConcatenate,
+    CutMix,
+    DynamicBucketingSampler,
+    K2SpeechRecognitionDataset,
+    PrecomputedFeatures,
+    SingleCutSampler,
+    SpecAugment,
+)
+from lhotse.dataset.input_strategies import OnTheFlyFeatures
+from lhotse.utils import fix_random_seed
+from torch.utils.data import DataLoader
+
+from icefall.utils import str2bool
+
+
+class _SeedWorkers:
+    def __init__(self, seed: int):
+        self.seed = seed
+
+    def __call__(self, worker_id: int):
+        fix_random_seed(self.seed + worker_id)
+
+
+class GigaSpeechAsrDataModule:
+    """
+    DataModule for k2 ASR experiments.
+    It assumes there is always one train and valid dataloader,
+    but there can be multiple test dataloaders (e.g. LibriSpeech test-clean
+    and test-other).
+
+    It contains all the common data pipeline modules used in ASR
+    experiments, e.g.:
+    - dynamic batch size,
+    - bucketing samplers,
+    - cut concatenation,
+    - augmentation,
+    - on-the-fly feature extraction
+
+    This class should be derived for specific corpora used in ASR tasks.
+    """
+
+    def __init__(self, args: argparse.Namespace):
+        self.args = args
+
+    @classmethod
+    def add_arguments(cls, parser: argparse.ArgumentParser):
+        group = parser.add_argument_group(
+            title="ASR data related options",
+            description="These options are used for the preparation of "
+            "PyTorch DataLoaders from Lhotse CutSet's -- they control the "
+            "effective batch sizes, sampling strategies, applied data "
+            "augmentations, etc.",
+        )
+        group.add_argument(
+            "--manifest-dir",
+            type=Path,
+            default=Path("data/fbank"),
+            help="Path to directory with train/valid/test cuts.",
+        )
+        group.add_argument(
+            "--max-duration",
+            type=int,
+            default=200.0,
+            help="Maximum pooled recordings duration (seconds) in a "
+            "single batch. You can reduce it if it causes CUDA OOM.",
+        )
+        group.add_argument(
+            "--bucketing-sampler",
+            type=str2bool,
+            default=True,
+            help="When enabled, the batches will come from buckets of "
+            "similar duration (saves padding frames).",
+        )
+        group.add_argument(
+            "--num-buckets",
+            type=int,
+            default=30,
+            help="The number of buckets for the DynamicBucketingSampler"
+            "(you might want to increase it for larger datasets).",
+        )
+        group.add_argument(
+            "--concatenate-cuts",
+            type=str2bool,
+            default=False,
+            help="When enabled, utterances (cuts) will be concatenated "
+            "to minimize the amount of padding.",
+        )
+        group.add_argument(
+            "--duration-factor",
+            type=float,
+            default=1.0,
+            help="Determines the maximum duration of a concatenated cut "
+            "relative to the duration of the longest cut in a batch.",
+        )
+        group.add_argument(
+            "--gap",
+            type=float,
+            default=1.0,
+            help="The amount of padding (in seconds) inserted between "
+            "concatenated cuts. This padding is filled with noise when "
+            "noise augmentation is used.",
+        )
+        group.add_argument(
+            "--on-the-fly-feats",
+            type=str2bool,
+            default=False,
+            help="When enabled, use on-the-fly cut mixing and feature "
+            "extraction. Will drop existing precomputed feature manifests "
+            "if available.",
+        )
+        group.add_argument(
+            "--shuffle",
+            type=str2bool,
+            default=True,
+            help="When enabled (=default), the examples will be "
+            "shuffled for each epoch.",
+        )
+        group.add_argument(
+            "--return-cuts",
+            type=str2bool,
+            default=True,
+            help="When enabled, each batch will have the "
+            "field: batch['supervisions']['cut'] with the cuts that "
+            "were used to construct it.",
+        )
+
+        group.add_argument(
+            "--num-workers",
+            type=int,
+            default=2,
+            help="The number of training dataloader workers that "
+            "collect the batches.",
+        )
+
+        group.add_argument(
+            "--enable-spec-aug",
+            type=str2bool,
+            default=True,
+            help="When enabled, use SpecAugment for training dataset.",
+        )
+
+        group.add_argument(
+            "--spec-aug-time-warp-factor",
+            type=int,
+            default=80,
+            help="Used only when --enable-spec-aug is True. "
+            "It specifies the factor for time warping in SpecAugment. "
+            "Larger values mean more warping. "
+            "A value less than 1 means to disable time warp.",
+        )
+
+        group.add_argument(
+            "--enable-musan",
+            type=str2bool,
+            default=True,
+            help="When enabled, select noise from MUSAN and mix it "
+            "with training dataset. ",
+        )
+
+        # GigaSpeech specific arguments
+        group.add_argument(
+            "--subset",
+            type=str,
+            default="XL",
+            help="Select the GigaSpeech subset (XS|S|M|L|XL)",
+        )
+        group.add_argument(
+            "--small-dev",
+            type=str2bool,
+            default=False,
+            help="Should we use only 1000 utterances for dev "
+            "(speeds up training)",
+        )
+
+    def train_dataloaders(
+        self,
+        cuts_train: CutSet,
+        sampler_state_dict: Optional[Dict[str, Any]] = None,
+    ) -> DataLoader:
+        """
+        Args:
+          cuts_train:
+            CutSet for training.
+          sampler_state_dict:
+            The state dict for the training sampler.
+        """
+
+        transforms = []
+        if self.args.enable_musan:
+            logging.info("Enable MUSAN")
+            logging.info("About to get Musan cuts")
+            cuts_musan = load_manifest(
+                self.args.manifest_dir / "cuts_musan.json.gz"
+            )
+            transforms.append(
+                CutMix(
+                    cuts=cuts_musan, prob=0.5, snr=(10, 20), preserve_id=True
+                )
+            )
+        else:
+            logging.info("Disable MUSAN")
+
+        if self.args.concatenate_cuts:
+            logging.info(
+                f"Using cut concatenation with duration factor "
+                f"{self.args.duration_factor} and gap {self.args.gap}."
+            )
+            # Cut concatenation should be the first transform in the list,
+            # so that if we e.g. mix noise in, it will fill the gaps between
+            # different utterances.
+            transforms = [
+                CutConcatenate(
+                    duration_factor=self.args.duration_factor, gap=self.args.gap
+                )
+            ] + transforms
+
+        input_transforms = []
+        if self.args.enable_spec_aug:
+            logging.info("Enable SpecAugment")
+            logging.info(
+                f"Time warp factor: {self.args.spec_aug_time_warp_factor}"
+            )
+            # Set the value of num_frame_masks according to Lhotse's version.
+            # In different Lhotse's versions, the default of num_frame_masks is
+            # different.
+            num_frame_masks = 10
+            num_frame_masks_parameter = inspect.signature(
+                SpecAugment.__init__
+            ).parameters["num_frame_masks"]
+            if num_frame_masks_parameter.default == 1:
+                num_frame_masks = 2
+            logging.info(f"Num frame mask: {num_frame_masks}")
+            input_transforms.append(
+                SpecAugment(
+                    time_warp_factor=self.args.spec_aug_time_warp_factor,
+                    num_frame_masks=num_frame_masks,
+                    features_mask_size=27,
+                    num_feature_masks=2,
+                    frames_mask_size=100,
+                )
+            )
+        else:
+            logging.info("Disable SpecAugment")
+
+        logging.info("About to create train dataset")
+        train = K2SpeechRecognitionDataset(
+            cut_transforms=transforms,
+            input_transforms=input_transforms,
+            return_cuts=self.args.return_cuts,
+        )
+
+        if self.args.on_the_fly_feats:
+            # NOTE: the PerturbSpeed transform should be added only if we
+            # remove it from data prep stage.
+            # Add on-the-fly speed perturbation; since originally it would
+            # have increased epoch size by 3, we will apply prob 2/3 and use
+            # 3x more epochs.
+            # Speed perturbation probably should come first before
+            # concatenation, but in principle the transforms order doesn't have
+            # to be strict (e.g. could be randomized)
+            # transforms = [PerturbSpeed(factors=[0.9, 1.1], p=2/3)] + transforms   # noqa
+            # Drop feats to be on the safe side.
+            train = K2SpeechRecognitionDataset(
+                cut_transforms=transforms,
+                input_strategy=OnTheFlyFeatures(
+                    Fbank(FbankConfig(num_mel_bins=80))
+                ),
+                input_transforms=input_transforms,
+                return_cuts=self.args.return_cuts,
+            )
+
+        if self.args.bucketing_sampler:
+            logging.info("Using DynamicBucketingSampler.")
+            train_sampler = DynamicBucketingSampler(
+                cuts_train,
+                max_duration=self.args.max_duration,
+                shuffle=self.args.shuffle,
+                num_buckets=self.args.num_buckets,
+                drop_last=True,
+            )
+        else:
+            logging.info("Using SingleCutSampler.")
+            train_sampler = SingleCutSampler(
+                cuts_train,
+                max_duration=self.args.max_duration,
+                shuffle=self.args.shuffle,
+            )
+        logging.info("About to create train dataloader")
+
+        if sampler_state_dict is not None:
+            logging.info("Loading sampler state dict")
+            train_sampler.load_state_dict(sampler_state_dict)
+
+        # 'seed' is derived from the current random state, which will have
+        # previously been set in the main process.
+        seed = torch.randint(0, 100000, ()).item()
+        worker_init_fn = _SeedWorkers(seed)
+
+        train_dl = DataLoader(
+            train,
+            sampler=train_sampler,
+            batch_size=None,
+            num_workers=self.args.num_workers,
+            persistent_workers=False,
+            worker_init_fn=worker_init_fn,
+        )
+
+        return train_dl
+
+    def valid_dataloaders(self, cuts_valid: CutSet) -> DataLoader:
+        transforms = []
+        if self.args.concatenate_cuts:
+            transforms = [
+                CutConcatenate(
+                    duration_factor=self.args.duration_factor, gap=self.args.gap
+                )
+            ] + transforms
+
+        logging.info("About to create dev dataset")
+        if self.args.on_the_fly_feats:
+            validate = K2SpeechRecognitionDataset(
+                cut_transforms=transforms,
+                input_strategy=OnTheFlyFeatures(
+                    Fbank(FbankConfig(num_mel_bins=80))
+                ),
+                return_cuts=self.args.return_cuts,
+            )
+        else:
+            validate = K2SpeechRecognitionDataset(
+                cut_transforms=transforms,
+                return_cuts=self.args.return_cuts,
+            )
+        valid_sampler = BucketingSampler(
+            cuts_valid,
+            max_duration=self.args.max_duration,
+            shuffle=False,
+        )
+        logging.info("About to create dev dataloader")
+        valid_dl = DataLoader(
+            validate,
+            sampler=valid_sampler,
+            batch_size=None,
+            num_workers=2,
+            persistent_workers=False,
+        )
+
+        return valid_dl
+
+    def test_dataloaders(self, cuts: CutSet) -> DataLoader:
+        logging.debug("About to create test dataset")
+        test = K2SpeechRecognitionDataset(
+            input_strategy=OnTheFlyFeatures(Fbank(FbankConfig(num_mel_bins=80)))
+            if self.args.on_the_fly_feats
+            else PrecomputedFeatures(),
+            return_cuts=self.args.return_cuts,
+        )
+        sampler = BucketingSampler(
+            cuts, max_duration=self.args.max_duration, shuffle=False
+        )
+        logging.debug("About to create test dataloader")
+        test_dl = DataLoader(
+            test,
+            batch_size=None,
+            sampler=sampler,
+            num_workers=self.args.num_workers,
+        )
+        return test_dl
+
+    @lru_cache()
+    def train_cuts(self) -> CutSet:
+        logging.info(f"About to get train_{self.args.subset} cuts")
+        path = self.args.manifest_dir / f"cuts_{self.args.subset}.jsonl.gz"
+        cuts_train = CutSet.from_jsonl_lazy(path)
+        return cuts_train
+
+    @lru_cache()
+    def dev_cuts(self) -> CutSet:
+        logging.info("About to get dev cuts")
+        cuts_valid = load_manifest(self.args.manifest_dir / "cuts_DEV.jsonl.gz")
+        if self.args.small_dev:
+            return cuts_valid.subset(first=1000)
+        else:
+            return cuts_valid
+
+    @lru_cache()
+    def test_cuts(self) -> CutSet:
+        logging.info("About to get test cuts")
+        return load_manifest(self.args.manifest_dir / "cuts_TEST.jsonl.gz")

egs/gigaspeech/ASR/pruned_transducer_stateless2/beam_search.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/beam_search.py

egs/gigaspeech/ASR/pruned_transducer_stateless2/conformer.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/conformer.py

egs/gigaspeech/ASR/pruned_transducer_stateless2/decode.py

@@ -0,0 +1,577 @@
+#!/usr/bin/env python3
+#
+# Copyright 2021 Xiaomi Corporation (Author: Fangjun Kuang)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+Usage:
+(1) greedy search
+./pruned_transducer_stateless2/decode.py \
+        --epoch 28 \
+        --avg 15 \
+        --exp-dir ./pruned_transducer_stateless2/exp \
+        --max-duration 600 \
+        --decoding-method greedy_search
+
+(2) beam search
+./pruned_transducer_stateless2/decode.py \
+        --epoch 28 \
+        --avg 15 \
+        --exp-dir ./pruned_transducer_stateless2/exp \
+        --max-duration 600 \
+        --decoding-method beam_search \
+        --beam-size 4
+
+(3) modified beam search
+./pruned_transducer_stateless2/decode.py \
+        --epoch 28 \
+        --avg 15 \
+        --exp-dir ./pruned_transducer_stateless2/exp \
+        --max-duration 600 \
+        --decoding-method modified_beam_search \
+        --beam-size 4
+
+(4) fast beam search
+./pruned_transducer_stateless2/decode.py \
+        --epoch 28 \
+        --avg 15 \
+        --exp-dir ./pruned_transducer_stateless2/exp \
+        --max-duration 600 \
+        --decoding-method fast_beam_search \
+        --beam 4 \
+        --max-contexts 4 \
+        --max-states 8
+"""
+
+
+import argparse
+import logging
+from collections import defaultdict
+from pathlib import Path
+from typing import Dict, List, Optional, Tuple
+
+import k2
+import sentencepiece as spm
+import torch
+import torch.nn as nn
+from asr_datamodule import GigaSpeechAsrDataModule
+from beam_search import (
+    beam_search,
+    fast_beam_search_one_best,
+    greedy_search,
+    greedy_search_batch,
+    modified_beam_search,
+)
+from gigaspeech_scoring import asr_text_post_processing
+from train import get_params, get_transducer_model
+
+from icefall.checkpoint import (
+    average_checkpoints,
+    find_checkpoints,
+    load_checkpoint,
+)
+from icefall.utils import (
+    AttributeDict,
+    setup_logger,
+    store_transcripts,
+    write_error_stats,
+)
+
+
+def get_parser():
+    parser = argparse.ArgumentParser(
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter
+    )
+
+    parser.add_argument(
+        "--epoch",
+        type=int,
+        default=29,
+        help="""It specifies the checkpoint to use for decoding.
+        Note: Epoch counts from 0.
+        You can specify --avg to use more checkpoints for model averaging.""",
+    )
+
+    parser.add_argument(
+        "--iter",
+        type=int,
+        default=0,
+        help="""If positive, --epoch is ignored and it
+        will use the checkpoint exp_dir/checkpoint-iter.pt.
+        You can specify --avg to use more checkpoints for model averaging.
+        """,
+    )
+
+    parser.add_argument(
+        "--avg",
+        type=int,
+        default=8,
+        help="Number of checkpoints to average. Automatically select "
+        "consecutive checkpoints before the checkpoint specified by "
+        "'--epoch' and '--iter'",
+    )
+
+    parser.add_argument(
+        "--exp-dir",
+        type=str,
+        default="pruned_transducer_stateless2/exp",
+        help="The experiment dir",
+    )
+
+    parser.add_argument(
+        "--bpe-model",
+        type=str,
+        default="data/lang_bpe_500/bpe.model",
+        help="Path to the BPE model",
+    )
+
+    parser.add_argument(
+        "--decoding-method",
+        type=str,
+        default="greedy_search",
+        help="""Possible values are:
+          - greedy_search
+          - beam_search
+          - modified_beam_search
+          - fast_beam_search
+        """,
+    )
+
+    parser.add_argument(
+        "--beam-size",
+        type=int,
+        default=4,
+        help="""An integer indicating how many candidates we will keep for each
+        frame. Used only when --decoding-method is beam_search or
+        modified_beam_search.""",
+    )
+
+    parser.add_argument(
+        "--beam",
+        type=float,
+        default=4,
+        help="""A floating point value to calculate the cutoff score during beam
+        search (i.e., `cutoff = max-score - beam`), which is the same as the
+        `beam` in Kaldi.
+        Used only when --decoding-method is fast_beam_search""",
+    )
+
+    parser.add_argument(
+        "--max-contexts",
+        type=int,
+        default=4,
+        help="""Used only when --decoding-method is
+        fast_beam_search""",
+    )
+
+    parser.add_argument(
+        "--max-states",
+        type=int,
+        default=8,
+        help="""Used only when --decoding-method is
+        fast_beam_search""",
+    )
+
+    parser.add_argument(
+        "--context-size",
+        type=int,
+        default=2,
+        help="The context size in the decoder. 1 means bigram; "
+        "2 means tri-gram",
+    )
+    parser.add_argument(
+        "--max-sym-per-frame",
+        type=int,
+        default=1,
+        help="""Maximum number of symbols per frame.
+        Used only when --decoding_method is greedy_search""",
+    )
+
+    return parser
+
+
+def post_processing(
+    results: List[Tuple[List[str], List[str]]],
+) -> List[Tuple[List[str], List[str]]]:
+    new_results = []
+    for ref, hyp in results:
+        new_ref = asr_text_post_processing(" ".join(ref)).split()
+        new_hyp = asr_text_post_processing(" ".join(hyp)).split()
+        new_results.append((new_ref, new_hyp))
+    return new_results
+
+
+def decode_one_batch(
+    params: AttributeDict,
+    model: nn.Module,
+    sp: spm.SentencePieceProcessor,
+    batch: dict,
+    decoding_graph: Optional[k2.Fsa] = None,
+) -> Dict[str, List[List[str]]]:
+    """Decode one batch and return the result in a dict. The dict has the
+    following format:
+
+        - key: It indicates the setting used for decoding. For example,
+               if greedy_search is used, it would be "greedy_search"
+               If beam search with a beam size of 7 is used, it would be
+               "beam_7"
+        - value: It contains the decoding result. `len(value)` equals to
+                 batch size. `value[i]` is the decoding result for the i-th
+                 utterance in the given batch.
+    Args:
+      params:
+        It's the return value of :func:`get_params`.
+      model:
+        The neural model.
+      sp:
+        The BPE model.
+      batch:
+        It is the return value from iterating
+        `lhotse.dataset.K2SpeechRecognitionDataset`. See its documentation
+        for the format of the `batch`.
+      decoding_graph:
+        The decoding graph. Can be either a `k2.trivial_graph` or HLG, Used
+        only when --decoding_method is fast_beam_search.
+    Returns:
+      Return the decoding result. See above description for the format of
+      the returned dict.
+    """
+    device = model.device
+    feature = batch["inputs"]
+    assert feature.ndim == 3
+
+    feature = feature.to(device)
+    # at entry, feature is (N, T, C)
+
+    supervisions = batch["supervisions"]
+    feature_lens = supervisions["num_frames"].to(device)
+
+    encoder_out, encoder_out_lens = model.encoder(
+        x=feature, x_lens=feature_lens
+    )
+    hyps = []
+
+    if params.decoding_method == "fast_beam_search":
+        hyp_tokens = fast_beam_search_one_best(
+            model=model,
+            decoding_graph=decoding_graph,
+            encoder_out=encoder_out,
+            encoder_out_lens=encoder_out_lens,
+            beam=params.beam,
+            max_contexts=params.max_contexts,
+            max_states=params.max_states,
+        )
+        for hyp in sp.decode(hyp_tokens):
+            hyps.append(hyp.split())
+    elif (
+        params.decoding_method == "greedy_search"
+        and params.max_sym_per_frame == 1
+    ):
+        hyp_tokens = greedy_search_batch(
+            model=model,
+            encoder_out=encoder_out,
+            encoder_out_lens=encoder_out_lens,
+        )
+        for hyp in sp.decode(hyp_tokens):
+            hyps.append(hyp.split())
+    elif params.decoding_method == "modified_beam_search":
+        hyp_tokens = modified_beam_search(
+            model=model,
+            encoder_out=encoder_out,
+            encoder_out_lens=encoder_out_lens,
+            beam=params.beam_size,
+        )
+        for hyp in sp.decode(hyp_tokens):
+            hyps.append(hyp.split())
+    else:
+        batch_size = encoder_out.size(0)
+
+        for i in range(batch_size):
+            # fmt: off
+            encoder_out_i = encoder_out[i:i+1, :encoder_out_lens[i]]
+            # fmt: on
+            if params.decoding_method == "greedy_search":
+                hyp = greedy_search(
+                    model=model,
+                    encoder_out=encoder_out_i,
+                    max_sym_per_frame=params.max_sym_per_frame,
+                )
+            elif params.decoding_method == "beam_search":
+                hyp = beam_search(
+                    model=model,
+                    encoder_out=encoder_out_i,
+                    beam=params.beam_size,
+                )
+            else:
+                raise ValueError(
+                    f"Unsupported decoding method: {params.decoding_method}"
+                )
+            hyps.append(sp.decode(hyp).split())
+
+    if params.decoding_method == "greedy_search":
+        return {"greedy_search": hyps}
+    elif params.decoding_method == "fast_beam_search":
+        return {
+            (
+                f"beam_{params.beam}_"
+                f"max_contexts_{params.max_contexts}_"
+                f"max_states_{params.max_states}"
+            ): hyps
+        }
+    else:
+        return {f"beam_size_{params.beam_size}": hyps}
+
+
+def decode_dataset(
+    dl: torch.utils.data.DataLoader,
+    params: AttributeDict,
+    model: nn.Module,
+    sp: spm.SentencePieceProcessor,
+    decoding_graph: Optional[k2.Fsa] = None,
+) -> Dict[str, List[Tuple[List[str], List[str]]]]:
+    """Decode dataset.
+
+    Args:
+      dl:
+        PyTorch's dataloader containing the dataset to decode.
+      params:
+        It is returned by :func:`get_params`.
+      model:
+        The neural model.
+      sp:
+        The BPE model.
+      decoding_graph:
+        The decoding graph. Can be either a `k2.trivial_graph` or HLG, Used
+        only when --decoding_method is fast_beam_search.
+    Returns:
+      Return a dict, whose key may be "greedy_search" if greedy search
+      is used, or it may be "beam_7" if beam size of 7 is used.
+      Its value is a list of tuples. Each tuple contains two elements:
+      The first is the reference transcript, and the second is the
+      predicted result.
+    """
+    num_cuts = 0
+
+    try:
+        num_batches = len(dl)
+    except TypeError:
+        num_batches = "?"
+
+    log_interval = 20
+
+    results = defaultdict(list)
+    for batch_idx, batch in enumerate(dl):
+        texts = batch["supervisions"]["text"]
+
+        hyps_dict = decode_one_batch(
+            params=params,
+            model=model,
+            sp=sp,
+            decoding_graph=decoding_graph,
+            batch=batch,
+        )
+
+        for name, hyps in hyps_dict.items():
+            this_batch = []
+            assert len(hyps) == len(texts)
+            for hyp_words, ref_text in zip(hyps, texts):
+                ref_words = ref_text.split()
+                this_batch.append((ref_words, hyp_words))
+
+            results[name].extend(this_batch)
+
+        num_cuts += len(texts)
+
+        if batch_idx % log_interval == 0:
+            batch_str = f"{batch_idx}/{num_batches}"
+
+            logging.info(
+                f"batch {batch_str}, cuts processed until now is {num_cuts}"
+            )
+    return results
+
+
+def save_results(
+    params: AttributeDict,
+    test_set_name: str,
+    results_dict: Dict[str, List[Tuple[List[int], List[int]]]],
+):
+    test_set_wers = dict()
+    for key, results in results_dict.items():
+        recog_path = (
+            params.res_dir / f"recogs-{test_set_name}-{key}-{params.suffix}.txt"
+        )
+        results = post_processing(results)
+        store_transcripts(filename=recog_path, texts=results)
+        logging.info(f"The transcripts are stored in {recog_path}")
+
+        # The following prints out WERs, per-word error statistics and aligned
+        # ref/hyp pairs.
+        errs_filename = (
+            params.res_dir / f"errs-{test_set_name}-{key}-{params.suffix}.txt"
+        )
+        with open(errs_filename, "w") as f:
+            wer = write_error_stats(
+                f, f"{test_set_name}-{key}", results, enable_log=True
+            )
+            test_set_wers[key] = wer
+
+        logging.info("Wrote detailed error stats to {}".format(errs_filename))
+
+    test_set_wers = sorted(test_set_wers.items(), key=lambda x: x[1])
+    errs_info = (
+        params.res_dir
+        / f"wer-summary-{test_set_name}-{key}-{params.suffix}.txt"
+    )
+    with open(errs_info, "w") as f:
+        print("settings\tWER", file=f)
+        for key, val in test_set_wers:
+            print("{}\t{}".format(key, val), file=f)
+
+    s = "\nFor {}, WER of different settings are:\n".format(test_set_name)
+    note = "\tbest for {}".format(test_set_name)
+    for key, val in test_set_wers:
+        s += "{}\t{}{}\n".format(key, val, note)
+        note = ""
+    logging.info(s)
+
+
+@torch.no_grad()
+def main():
+    parser = get_parser()
+    GigaSpeechAsrDataModule.add_arguments(parser)
+    args = parser.parse_args()
+    args.exp_dir = Path(args.exp_dir)
+
+    params = get_params()
+    params.update(vars(args))
+
+    assert params.decoding_method in (
+        "greedy_search",
+        "beam_search",
+        "fast_beam_search",
+        "modified_beam_search",
+    )
+    params.res_dir = params.exp_dir / params.decoding_method
+
+    if params.iter > 0:
+        params.suffix = f"iter-{params.iter}-avg-{params.avg}"
+    else:
+        params.suffix = f"epoch-{params.epoch}-avg-{params.avg}"
+
+    if "fast_beam_search" in params.decoding_method:
+        params.suffix += f"-beam-{params.beam}"
+        params.suffix += f"-max-contexts-{params.max_contexts}"
+        params.suffix += f"-max-states-{params.max_states}"
+    elif "beam_search" in params.decoding_method:
+        params.suffix += f"-beam-{params.beam_size}"
+    else:
+        params.suffix += f"-context-{params.context_size}"
+        params.suffix += f"-max-sym-per-frame-{params.max_sym_per_frame}"
+
+    setup_logger(f"{params.res_dir}/log-decode-{params.suffix}")
+    logging.info("Decoding started")
+
+    device = torch.device("cpu")
+    if torch.cuda.is_available():
+        device = torch.device("cuda", 0)
+
+    logging.info(f"Device: {device}")
+
+    sp = spm.SentencePieceProcessor()
+    sp.load(params.bpe_model)
+
+    # <blk> and <unk> is defined in local/train_bpe_model.py
+    params.blank_id = sp.piece_to_id("<blk>")
+    params.unk_id = sp.piece_to_id("<unk>")
+    params.vocab_size = sp.get_piece_size()
+
+    logging.info(params)
+
+    logging.info("About to create model")
+    model = get_transducer_model(params)
+
+    if params.iter > 0:
+        filenames = find_checkpoints(params.exp_dir, iteration=-params.iter)[
+            : params.avg
+        ]
+        if len(filenames) == 0:
+            raise ValueError(
+                f"No checkpoints found for"
+                f" --iter {params.iter}, --avg {params.avg}"
+            )
+        elif len(filenames) < params.avg:
+            raise ValueError(
+                f"Not enough checkpoints ({len(filenames)}) found for"
+                f" --iter {params.iter}, --avg {params.avg}"
+            )
+        logging.info(f"averaging {filenames}")
+        model.to(device)
+        model.load_state_dict(average_checkpoints(filenames, device=device))
+    elif params.avg == 1:
+        load_checkpoint(f"{params.exp_dir}/epoch-{params.epoch}.pt", model)
+    else:
+        start = params.epoch - params.avg + 1
+        filenames = []
+        for i in range(start, params.epoch + 1):
+            if start >= 0:
+                filenames.append(f"{params.exp_dir}/epoch-{i}.pt")
+        logging.info(f"averaging {filenames}")
+        model.to(device)
+        model.load_state_dict(average_checkpoints(filenames, device=device))
+
+    model.to(device)
+    model.eval()
+    model.device = device
+
+    if params.decoding_method == "fast_beam_search":
+        decoding_graph = k2.trivial_graph(params.vocab_size - 1, device=device)
+    else:
+        decoding_graph = None
+
+    num_param = sum([p.numel() for p in model.parameters()])
+    logging.info(f"Number of model parameters: {num_param}")
+
+    gigaspeech = GigaSpeechAsrDataModule(args)
+
+    dev_cuts = gigaspeech.dev_cuts()
+    test_cuts = gigaspeech.test_cuts()
+
+    dev_dl = gigaspeech.test_dataloaders(dev_cuts)
+    test_dl = gigaspeech.test_dataloaders(test_cuts)
+
+    test_sets = ["dev", "test"]
+    test_dls = [dev_dl, test_dl]
+
+    for test_set, test_dl in zip(test_sets, test_dls):
+        results_dict = decode_dataset(
+            dl=test_dl,
+            params=params,
+            model=model,
+            sp=sp,
+            decoding_graph=decoding_graph,
+        )
+
+        save_results(
+            params=params,
+            test_set_name=test_set,
+            results_dict=results_dict,
+        )
+
+    logging.info("Done!")
+
+
+if __name__ == "__main__":
+    main()

egs/gigaspeech/ASR/pruned_transducer_stateless2/decoder.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/decoder.py

egs/gigaspeech/ASR/pruned_transducer_stateless2/encoder_interface.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/encoder_interface.py

egs/gigaspeech/ASR/pruned_transducer_stateless2/export.py

@@ -0,0 +1,214 @@
+#!/usr/bin/env python3
+#
+# Copyright 2021 Xiaomi Corporation (Author: Fangjun Kuang)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# This script converts several saved checkpoints
+# to a single one using model averaging.
+"""
+Usage:
+./pruned_transducer_stateless2/export.py \
+  --exp-dir ./pruned_transducer_stateless2/exp \
+  --bpe-model data/lang_bpe_500/bpe.model \
+  --epoch 20 \
+  --avg 10
+
+It will generate a file exp_dir/pretrained.pt
+
+To use the generated file with `pruned_transducer_stateless2/decode.py`,
+you can do:
+
+    cd /path/to/exp_dir
+    ln -s pretrained.pt epoch-9999.pt
+
+    cd /path/to/egs/librispeech/ASR
+    ./pruned_transducer_stateless2/decode.py \
+        --exp-dir ./pruned_transducer_stateless2/exp \
+        --epoch 9999 \
+        --avg 1 \
+        --max-duration 100 \
+        --bpe-model data/lang_bpe_500/bpe.model
+"""
+
+import argparse
+import logging
+from pathlib import Path
+
+import sentencepiece as spm
+import torch
+from train import get_params, get_transducer_model
+
+from icefall.checkpoint import (
+    average_checkpoints,
+    find_checkpoints,
+    load_checkpoint,
+)
+from icefall.utils import str2bool
+
+
+def get_parser():
+    parser = argparse.ArgumentParser(
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter
+    )
+
+    parser.add_argument(
+        "--epoch",
+        type=int,
+        default=28,
+        help="""It specifies the checkpoint to use for averaging.
+        Note: Epoch counts from 0.
+        You can specify --avg to use more checkpoints for model averaging.""",
+    )
+
+    parser.add_argument(
+        "--iter",
+        type=int,
+        default=0,
+        help="""If positive, --epoch is ignored and it
+        will use the checkpoint exp_dir/checkpoint-iter.pt.
+        You can specify --avg to use more checkpoints for model averaging.
+        """,
+    )
+
+    parser.add_argument(
+        "--avg",
+        type=int,
+        default=15,
+        help="Number of checkpoints to average. Automatically select "
+        "consecutive checkpoints before the checkpoint specified by "
+        "'--epoch' and '--iter'",
+    )
+
+    parser.add_argument(
+        "--exp-dir",
+        type=str,
+        default="pruned_transducer_stateless2/exp",
+        help="""It specifies the directory where all training related
+        files, e.g., checkpoints, log, etc, are saved
+        """,
+    )
+
+    parser.add_argument(
+        "--bpe-model",
+        type=str,
+        default="data/lang_bpe_500/bpe.model",
+        help="Path to the BPE model",
+    )
+
+    parser.add_argument(
+        "--jit",
+        type=str2bool,
+        default=False,
+        help="""True to save a model after applying torch.jit.script.
+        """,
+    )
+
+    parser.add_argument(
+        "--context-size",
+        type=int,
+        default=2,
+        help="The context size in the decoder. 1 means bigram; "
+        "2 means tri-gram",
+    )
+
+    return parser
+
+
+def main():
+    args = get_parser().parse_args()
+    args.exp_dir = Path(args.exp_dir)
+
+    assert args.jit is False, "Support torchscript will be added later"
+
+    params = get_params()
+    params.update(vars(args))
+
+    device = torch.device("cpu")
+    if torch.cuda.is_available():
+        device = torch.device("cuda", 0)
+
+    logging.info(f"device: {device}")
+
+    sp = spm.SentencePieceProcessor()
+    sp.load(params.bpe_model)
+
+    # <blk> is defined in local/train_bpe_model.py
+    params.blank_id = sp.piece_to_id("<blk>")
+    params.vocab_size = sp.get_piece_size()
+
+    logging.info(params)
+
+    logging.info("About to create model")
+    model = get_transducer_model(params)
+
+    model.to(device)
+
+    if params.iter > 0:
+        filenames = find_checkpoints(params.exp_dir, iteration=-params.iter)[
+            : params.avg
+        ]
+        if len(filenames) == 0:
+            raise ValueError(
+                f"No checkpoints found for"
+                f" --iter {params.iter}, --avg {params.avg}"
+            )
+        elif len(filenames) < params.avg:
+            raise ValueError(
+                f"Not enough checkpoints ({len(filenames)}) found for"
+                f" --iter {params.iter}, --avg {params.avg}"
+            )
+        logging.info(f"averaging {filenames}")
+        model.to(device)
+        model.load_state_dict(average_checkpoints(filenames, device=device))
+    elif params.avg == 1:
+        load_checkpoint(f"{params.exp_dir}/epoch-{params.epoch}.pt", model)
+    else:
+        start = params.epoch - params.avg + 1
+        filenames = []
+        for i in range(start, params.epoch + 1):
+            if start >= 0:
+                filenames.append(f"{params.exp_dir}/epoch-{i}.pt")
+        logging.info(f"averaging {filenames}")
+        model.to(device)
+        model.load_state_dict(average_checkpoints(filenames, device=device))
+
+    model.eval()
+
+    model.to("cpu")
+    model.eval()
+
+    if params.jit:
+        logging.info("Using torch.jit.script")
+        model = torch.jit.script(model)
+        filename = params.exp_dir / "cpu_jit.pt"
+        model.save(str(filename))
+        logging.info(f"Saved to {filename}")
+    else:
+        logging.info("Not using torch.jit.script")
+        # Save it using a format so that it can be loaded
+        # by :func:`load_checkpoint`
+        filename = params.exp_dir / "pretrained.pt"
+        torch.save({"model": model.state_dict()}, str(filename))
+        logging.info(f"Saved to {filename}")
+
+
+if __name__ == "__main__":
+    formatter = (
+        "%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s"
+    )
+
+    logging.basicConfig(format=formatter, level=logging.INFO)
+    main()

egs/gigaspeech/ASR/pruned_transducer_stateless2/gigaspeech_scoring.py

@@ -0,0 +1 @@
+../conformer_ctc/gigaspeech_scoring.py

egs/gigaspeech/ASR/pruned_transducer_stateless2/joiner.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/joiner.py

egs/gigaspeech/ASR/pruned_transducer_stateless2/model.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/model.py

egs/gigaspeech/ASR/pruned_transducer_stateless2/optim.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/optim.py

egs/gigaspeech/ASR/pruned_transducer_stateless2/scaling.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/scaling.py

egs/gigaspeech/ASR/pruned_transducer_stateless2/train.py

@@ -0,0 +1,974 @@
+#!/usr/bin/env python3
+# Copyright    2021  Xiaomi Corp.        (authors: Fangjun Kuang,
+#                                                  Wei Kang
+#                                                  Mingshuang Luo)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+Usage:
+
+export CUDA_VISIBLE_DEVICES="0,1,2,3"
+
+./pruned_transducer_stateless2/train.py \
+  --world-size 4 \
+  --num-epochs 30 \
+  --start-epoch 0 \
+  --exp-dir pruned_transducer_stateless2/exp \
+  --full-libri 1 \
+  --max-duration 300
+
+# For mix precision training:
+
+./pruned_transducer_stateless2/train.py \
+  --world-size 4 \
+  --num-epochs 30 \
+  --start-epoch 0 \
+  --use_fp16 1 \
+  --exp-dir pruned_transducer_stateless2/exp \
+  --full-libri 1 \
+  --max-duration 550
+
+"""
+
+
+import argparse
+import logging
+import warnings
+from pathlib import Path
+from shutil import copyfile
+from typing import Any, Dict, Optional, Tuple, Union
+
+import k2
+import optim
+import sentencepiece as spm
+import torch
+import torch.multiprocessing as mp
+import torch.nn as nn
+from asr_datamodule import GigaSpeechAsrDataModule
+from conformer import Conformer
+from decoder import Decoder
+from joiner import Joiner
+from lhotse.dataset.sampling.base import CutSampler
+from lhotse.utils import fix_random_seed
+from model import Transducer
+from optim import Eden, Eve
+from torch import Tensor
+from torch.cuda.amp import GradScaler
+from torch.nn.parallel import DistributedDataParallel as DDP
+from torch.utils.tensorboard import SummaryWriter
+
+from icefall import diagnostics
+from icefall.checkpoint import load_checkpoint, remove_checkpoints
+from icefall.checkpoint import save_checkpoint as save_checkpoint_impl
+from icefall.checkpoint import save_checkpoint_with_global_batch_idx
+from icefall.dist import cleanup_dist, setup_dist
+from icefall.env import get_env_info
+from icefall.utils import AttributeDict, MetricsTracker, setup_logger, str2bool
+
+LRSchedulerType = Union[
+    torch.optim.lr_scheduler._LRScheduler, optim.LRScheduler
+]
+
+
+def get_parser():
+    parser = argparse.ArgumentParser(
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter
+    )
+
+    parser.add_argument(
+        "--world-size",
+        type=int,
+        default=1,
+        help="Number of GPUs for DDP training.",
+    )
+
+    parser.add_argument(
+        "--master-port",
+        type=int,
+        default=12354,
+        help="Master port to use for DDP training.",
+    )
+
+    parser.add_argument(
+        "--tensorboard",
+        type=str2bool,
+        default=True,
+        help="Should various information be logged in tensorboard.",
+    )
+
+    parser.add_argument(
+        "--num-epochs",
+        type=int,
+        default=30,
+        help="Number of epochs to train.",
+    )
+
+    parser.add_argument(
+        "--start-epoch",
+        type=int,
+        default=0,
+        help="""Resume training from from this epoch.
+        If it is positive, it will load checkpoint from
+        transducer_stateless2/exp/epoch-{start_epoch-1}.pt
+        """,
+    )
+
+    parser.add_argument(
+        "--start-batch",
+        type=int,
+        default=0,
+        help="""If positive, --start-epoch is ignored and
+        it loads the checkpoint from exp-dir/checkpoint-{start_batch}.pt
+        """,
+    )
+
+    parser.add_argument(
+        "--exp-dir",
+        type=str,
+        default="pruned_transducer_stateless2/exp",
+        help="""The experiment dir.
+        It specifies the directory where all training related
+        files, e.g., checkpoints, log, etc, are saved
+        """,
+    )
+
+    parser.add_argument(
+        "--bpe-model",
+        type=str,
+        default="data/lang_bpe_500/bpe.model",
+        help="Path to the BPE model",
+    )
+
+    parser.add_argument(
+        "--initial-lr",
+        type=float,
+        default=0.003,
+        help="The initial learning rate.  This value should not need to be changed.",
+    )
+
+    parser.add_argument(
+        "--lr-batches",
+        type=float,
+        default=5000,
+        help="""Number of steps that affects how rapidly the learning rate decreases.
+        We suggest not to change this.""",
+    )
+
+    parser.add_argument(
+        "--lr-epochs",
+        type=float,
+        default=6,
+        help="""Number of epochs that affects how rapidly the learning rate decreases.
+        """,
+    )
+
+    parser.add_argument(
+        "--context-size",
+        type=int,
+        default=2,
+        help="The context size in the decoder. 1 means bigram; "
+        "2 means tri-gram",
+    )
+
+    parser.add_argument(
+        "--prune-range",
+        type=int,
+        default=5,
+        help="The prune range for rnnt loss, it means how many symbols(context)"
+        "we are using to compute the loss",
+    )
+
+    parser.add_argument(
+        "--lm-scale",
+        type=float,
+        default=0.25,
+        help="The scale to smooth the loss with lm "
+        "(output of prediction network) part.",
+    )
+
+    parser.add_argument(
+        "--am-scale",
+        type=float,
+        default=0.0,
+        help="The scale to smooth the loss with am (output of encoder network)"
+        "part.",
+    )
+
+    parser.add_argument(
+        "--simple-loss-scale",
+        type=float,
+        default=0.5,
+        help="To get pruning ranges, we will calculate a simple version"
+        "loss(joiner is just addition), this simple loss also uses for"
+        "training (as a regularization item). We will scale the simple loss"
+        "with this parameter before adding to the final loss.",
+    )
+
+    parser.add_argument(
+        "--seed",
+        type=int,
+        default=42,
+        help="The seed for random generators intended for reproducibility",
+    )
+
+    parser.add_argument(
+        "--print-diagnostics",
+        type=str2bool,
+        default=False,
+        help="Accumulate stats on activations, print them and exit.",
+    )
+
+    parser.add_argument(
+        "--save-every-n",
+        type=int,
+        default=8000,
+        help="""Save checkpoint after processing this number of batches"
+        periodically. We save checkpoint to exp-dir/ whenever
+        params.batch_idx_train % save_every_n == 0. The checkpoint filename
+        has the form: f'exp-dir/checkpoint-{params.batch_idx_train}.pt'
+        Note: It also saves checkpoint to `exp-dir/epoch-xxx.pt` at the
+        end of each epoch where `xxx` is the epoch number counting from 0.
+        """,
+    )
+
+    parser.add_argument(
+        "--keep-last-k",
+        type=int,
+        default=20,
+        help="""Only keep this number of checkpoints on disk.
+        For instance, if it is 3, there are only 3 checkpoints
+        in the exp-dir with filenames `checkpoint-xxx.pt`.
+        It does not affect checkpoints with name `epoch-xxx.pt`.
+        """,
+    )
+
+    parser.add_argument(
+        "--use-fp16",
+        type=str2bool,
+        default=False,
+        help="Whether to use half precision training.",
+    )
+
+    return parser
+
+
+def get_params() -> AttributeDict:
+    """Return a dict containing training parameters.
+
+    All training related parameters that are not passed from the commandline
+    are saved in the variable `params`.
+
+    Commandline options are merged into `params` after they are parsed, so
+    you can also access them via `params`.
+
+    Explanation of options saved in `params`:
+
+        - best_train_loss: Best training loss so far. It is used to select
+                           the model that has the lowest training loss. It is
+                           updated during the training.
+
+        - best_valid_loss: Best validation loss so far. It is used to select
+                           the model that has the lowest validation loss. It is
+                           updated during the training.
+
+        - best_train_epoch: It is the epoch that has the best training loss.
+
+        - best_valid_epoch: It is the epoch that has the best validation loss.
+
+        - batch_idx_train: Used to writing statistics to tensorboard. It
+                           contains number of batches trained so far across
+                           epochs.
+
+        - log_interval:  Print training loss if batch_idx % log_interval` is 0
+
+        - reset_interval: Reset statistics if batch_idx % reset_interval is 0
+
+        - valid_interval:  Run validation if batch_idx % valid_interval is 0
+
+        - feature_dim: The model input dim. It has to match the one used
+                       in computing features.
+
+        - subsampling_factor:  The subsampling factor for the model.
+
+        - encoder_dim: Hidden dim for multi-head attention model.
+
+        - num_decoder_layers: Number of decoder layer of transformer decoder.
+
+        - warm_step: The warm_step for Noam optimizer.
+    """
+    params = AttributeDict(
+        {
+            "best_train_loss": float("inf"),
+            "best_valid_loss": float("inf"),
+            "best_train_epoch": -1,
+            "best_valid_epoch": -1,
+            "batch_idx_train": 0,
+            "log_interval": 500,
+            "reset_interval": 2000,
+            "valid_interval": 20000,
+            # parameters for conformer
+            "feature_dim": 80,
+            "subsampling_factor": 4,
+            "encoder_dim": 512,
+            "nhead": 8,
+            "dim_feedforward": 2048,
+            "num_encoder_layers": 12,
+            # parameters for decoder
+            "decoder_dim": 512,
+            # parameters for joiner
+            "joiner_dim": 512,
+            # parameters for Noam
+            "model_warm_step": 20000,  # arg given to model, not for lrate
+            "env_info": get_env_info(),
+        }
+    )
+
+    return params
+
+
+def get_encoder_model(params: AttributeDict) -> nn.Module:
+    # TODO: We can add an option to switch between Conformer and Transformer
+    encoder = Conformer(
+        num_features=params.feature_dim,
+        subsampling_factor=params.subsampling_factor,
+        d_model=params.encoder_dim,
+        nhead=params.nhead,
+        dim_feedforward=params.dim_feedforward,
+        num_encoder_layers=params.num_encoder_layers,
+    )
+    return encoder
+
+
+def get_decoder_model(params: AttributeDict) -> nn.Module:
+    decoder = Decoder(
+        vocab_size=params.vocab_size,
+        decoder_dim=params.decoder_dim,
+        blank_id=params.blank_id,
+        context_size=params.context_size,
+    )
+    return decoder
+
+
+def get_joiner_model(params: AttributeDict) -> nn.Module:
+    joiner = Joiner(
+        encoder_dim=params.encoder_dim,
+        decoder_dim=params.decoder_dim,
+        joiner_dim=params.joiner_dim,
+        vocab_size=params.vocab_size,
+    )
+    return joiner
+
+
+def get_transducer_model(params: AttributeDict) -> nn.Module:
+    encoder = get_encoder_model(params)
+    decoder = get_decoder_model(params)
+    joiner = get_joiner_model(params)
+
+    model = Transducer(
+        encoder=encoder,
+        decoder=decoder,
+        joiner=joiner,
+        encoder_dim=params.encoder_dim,
+        decoder_dim=params.decoder_dim,
+        joiner_dim=params.joiner_dim,
+        vocab_size=params.vocab_size,
+    )
+    return model
+
+
+def load_checkpoint_if_available(
+    params: AttributeDict,
+    model: nn.Module,
+    optimizer: Optional[torch.optim.Optimizer] = None,
+    scheduler: Optional[LRSchedulerType] = None,
+) -> Optional[Dict[str, Any]]:
+    """Load checkpoint from file.
+
+    If params.start_batch is positive, it will load the checkpoint from
+    `params.exp_dir/checkpoint-{params.start_batch}.pt`. Otherwise, if
+    params.start_epoch is positive, it will load the checkpoint from
+    `params.start_epoch - 1`.
+
+    Apart from loading state dict for `model` and `optimizer` it also updates
+    `best_train_epoch`, `best_train_loss`, `best_valid_epoch`,
+    and `best_valid_loss` in `params`.
+
+    Args:
+      params:
+        The return value of :func:`get_params`.
+      model:
+        The training model.
+      optimizer:
+        The optimizer that we are using.
+      scheduler:
+        The scheduler that we are using.
+    Returns:
+      Return a dict containing previously saved training info.
+    """
+    if params.start_batch > 0:
+        filename = params.exp_dir / f"checkpoint-{params.start_batch}.pt"
+    elif params.start_epoch > 0:
+        filename = params.exp_dir / f"epoch-{params.start_epoch-1}.pt"
+    else:
+        return None
+
+    assert filename.is_file(), f"{filename} does not exist!"
+
+    saved_params = load_checkpoint(
+        filename,
+        model=model,
+        optimizer=optimizer,
+        scheduler=scheduler,
+    )
+
+    keys = [
+        "best_train_epoch",
+        "best_valid_epoch",
+        "batch_idx_train",
+        "best_train_loss",
+        "best_valid_loss",
+    ]
+    for k in keys:
+        params[k] = saved_params[k]
+
+    if params.start_batch > 0:
+        if "cur_epoch" in saved_params:
+            params["start_epoch"] = saved_params["cur_epoch"]
+
+        if "cur_batch_idx" in saved_params:
+            params["cur_batch_idx"] = saved_params["cur_batch_idx"]
+
+    return saved_params
+
+
+def save_checkpoint(
+    params: AttributeDict,
+    model: nn.Module,
+    optimizer: Optional[torch.optim.Optimizer] = None,
+    scheduler: Optional[LRSchedulerType] = None,
+    sampler: Optional[CutSampler] = None,
+    scaler: Optional[GradScaler] = None,
+    rank: int = 0,
+) -> None:
+    """Save model, optimizer, scheduler and training stats to file.
+
+    Args:
+      params:
+        It is returned by :func:`get_params`.
+      model:
+        The training model.
+      optimizer:
+        The optimizer used in the training.
+      sampler:
+       The sampler for the training dataset.
+      scaler:
+        The scaler used for mix precision training.
+    """
+    if rank != 0:
+        return
+    filename = params.exp_dir / f"epoch-{params.cur_epoch}.pt"
+    save_checkpoint_impl(
+        filename=filename,
+        model=model,
+        params=params,
+        optimizer=optimizer,
+        scheduler=scheduler,
+        sampler=sampler,
+        scaler=scaler,
+        rank=rank,
+    )
+
+    if params.best_train_epoch == params.cur_epoch:
+        best_train_filename = params.exp_dir / "best-train-loss.pt"
+        copyfile(src=filename, dst=best_train_filename)
+
+    if params.best_valid_epoch == params.cur_epoch:
+        best_valid_filename = params.exp_dir / "best-valid-loss.pt"
+        copyfile(src=filename, dst=best_valid_filename)
+
+
+def compute_loss(
+    params: AttributeDict,
+    model: nn.Module,
+    sp: spm.SentencePieceProcessor,
+    batch: dict,
+    is_training: bool,
+    warmup: float = 1.0,
+) -> Tuple[Tensor, MetricsTracker]:
+    """
+    Compute CTC loss given the model and its inputs.
+
+    Args:
+      params:
+        Parameters for training. See :func:`get_params`.
+      model:
+        The model for training. It is an instance of Conformer in our case.
+      batch:
+        A batch of data. See `lhotse.dataset.K2SpeechRecognitionDataset()`
+        for the content in it.
+      is_training:
+        True for training. False for validation. When it is True, this
+        function enables autograd during computation; when it is False, it
+        disables autograd.
+     warmup: a floating point value which increases throughout training;
+        values >= 1.0 are fully warmed up and have all modules present.
+    """
+    device = model.device
+    feature = batch["inputs"]
+    # at entry, feature is (N, T, C)
+    assert feature.ndim == 3
+    feature = feature.to(device)
+
+    supervisions = batch["supervisions"]
+    feature_lens = supervisions["num_frames"].to(device)
+
+    texts = batch["supervisions"]["text"]
+    y = sp.encode(texts, out_type=int)
+    y = k2.RaggedTensor(y).to(device)
+
+    with torch.set_grad_enabled(is_training):
+        simple_loss, pruned_loss = model(
+            x=feature,
+            x_lens=feature_lens,
+            y=y,
+            prune_range=params.prune_range,
+            am_scale=params.am_scale,
+            lm_scale=params.lm_scale,
+            warmup=warmup,
+        )
+        # after the main warmup step, we keep pruned_loss_scale small
+        # for the same amount of time (model_warm_step), to avoid
+        # overwhelming the simple_loss and causing it to diverge,
+        # in case it had not fully learned the alignment yet.
+        pruned_loss_scale = (
+            0.0
+            if warmup < 1.0
+            else (0.1 if warmup > 1.0 and warmup < 2.0 else 1.0)
+        )
+        loss = (
+            params.simple_loss_scale * simple_loss
+            + pruned_loss_scale * pruned_loss
+        )
+
+    assert loss.requires_grad == is_training
+
+    info = MetricsTracker()
+    with warnings.catch_warnings():
+        warnings.simplefilter("ignore")
+        info["frames"] = (
+            (feature_lens // params.subsampling_factor).sum().item()
+        )
+
+    # Note: We use reduction=sum while computing the loss.
+    info["loss"] = loss.detach().cpu().item()
+    info["simple_loss"] = simple_loss.detach().cpu().item()
+    info["pruned_loss"] = pruned_loss.detach().cpu().item()
+
+    return loss, info
+
+
+def compute_validation_loss(
+    params: AttributeDict,
+    model: nn.Module,
+    sp: spm.SentencePieceProcessor,
+    valid_dl: torch.utils.data.DataLoader,
+    world_size: int = 1,
+) -> MetricsTracker:
+    """Run the validation process."""
+    model.eval()
+
+    tot_loss = MetricsTracker()
+
+    for batch_idx, batch in enumerate(valid_dl):
+        loss, loss_info = compute_loss(
+            params=params,
+            model=model,
+            sp=sp,
+            batch=batch,
+            is_training=False,
+        )
+        assert loss.requires_grad is False
+        tot_loss = tot_loss + loss_info
+
+    if world_size > 1:
+        tot_loss.reduce(loss.device)
+
+    loss_value = tot_loss["loss"] / tot_loss["frames"]
+    if loss_value < params.best_valid_loss:
+        params.best_valid_epoch = params.cur_epoch
+        params.best_valid_loss = loss_value
+
+    return tot_loss
+
+
+def train_one_epoch(
+    params: AttributeDict,
+    model: nn.Module,
+    optimizer: torch.optim.Optimizer,
+    scheduler: LRSchedulerType,
+    sp: spm.SentencePieceProcessor,
+    train_dl: torch.utils.data.DataLoader,
+    valid_dl: torch.utils.data.DataLoader,
+    scaler: GradScaler,
+    tb_writer: Optional[SummaryWriter] = None,
+    world_size: int = 1,
+    rank: int = 0,
+) -> None:
+    """Train the model for one epoch.
+
+    The training loss from the mean of all frames is saved in
+    `params.train_loss`. It runs the validation process every
+    `params.valid_interval` batches.
+
+    Args:
+      params:
+        It is returned by :func:`get_params`.
+      model:
+        The model for training.
+      optimizer:
+        The optimizer we are using.
+      scheduler:
+        The learning rate scheduler, we call step() every step.
+      train_dl:
+        Dataloader for the training dataset.
+      valid_dl:
+        Dataloader for the validation dataset.
+      scaler:
+        The scaler used for mix precision training.
+      tb_writer:
+        Writer to write log messages to tensorboard.
+      world_size:
+        Number of nodes in DDP training. If it is 1, DDP is disabled.
+      rank:
+        The rank of the node in DDP training. If no DDP is used, it should
+        be set to 0.
+    """
+    model.train()
+
+    tot_loss = MetricsTracker()
+
+    cur_batch_idx = params.get("cur_batch_idx", 0)
+
+    for batch_idx, batch in enumerate(train_dl):
+        if batch_idx < cur_batch_idx:
+            continue
+        cur_batch_idx = batch_idx
+
+        params.batch_idx_train += 1
+        batch_size = len(batch["supervisions"]["text"])
+
+        with torch.cuda.amp.autocast(enabled=params.use_fp16):
+            loss, loss_info = compute_loss(
+                params=params,
+                model=model,
+                sp=sp,
+                batch=batch,
+                is_training=True,
+                warmup=(params.batch_idx_train / params.model_warm_step),
+            )
+        # summary stats
+        tot_loss = (tot_loss * (1 - 1 / params.reset_interval)) + loss_info
+
+        # NOTE: We use reduction==sum and loss is computed over utterances
+        # in the batch and there is no normalization to it so far.
+        scaler.scale(loss).backward()
+        scheduler.step_batch(params.batch_idx_train)
+        scaler.step(optimizer)
+        scaler.update()
+        optimizer.zero_grad()
+
+        if params.print_diagnostics and batch_idx == 30:
+            return
+
+        if (
+            params.batch_idx_train > 0
+            and params.batch_idx_train % params.save_every_n == 0
+        ):
+            params.cur_batch_idx = batch_idx
+            save_checkpoint_with_global_batch_idx(
+                out_dir=params.exp_dir,
+                global_batch_idx=params.batch_idx_train,
+                model=model,
+                params=params,
+                optimizer=optimizer,
+                scheduler=scheduler,
+                sampler=train_dl.sampler,
+                scaler=scaler,
+                rank=rank,
+            )
+            del params.cur_batch_idx
+            remove_checkpoints(
+                out_dir=params.exp_dir,
+                topk=params.keep_last_k,
+                rank=rank,
+            )
+
+        if batch_idx % params.log_interval == 0:
+            cur_lr = scheduler.get_last_lr()[0]
+            logging.info(
+                f"Epoch {params.cur_epoch}, "
+                f"batch {batch_idx}, loss[{loss_info}], "
+                f"tot_loss[{tot_loss}], batch size: {batch_size}, "
+                f"lr: {cur_lr:.2e}"
+            )
+
+            if tb_writer is not None:
+                tb_writer.add_scalar(
+                    "train/learning_rate", cur_lr, params.batch_idx_train
+                )
+
+                loss_info.write_summary(
+                    tb_writer, "train/current_", params.batch_idx_train
+                )
+                tot_loss.write_summary(
+                    tb_writer, "train/tot_", params.batch_idx_train
+                )
+
+        if batch_idx > 0 and batch_idx % params.valid_interval == 0:
+            logging.info("Computing validation loss")
+            valid_info = compute_validation_loss(
+                params=params,
+                model=model,
+                sp=sp,
+                valid_dl=valid_dl,
+                world_size=world_size,
+            )
+            model.train()
+            logging.info(f"Epoch {params.cur_epoch}, validation: {valid_info}")
+            if tb_writer is not None:
+                valid_info.write_summary(
+                    tb_writer, "train/valid_", params.batch_idx_train
+                )
+
+    loss_value = tot_loss["loss"] / tot_loss["frames"]
+    params.train_loss = loss_value
+    if params.train_loss < params.best_train_loss:
+        params.best_train_epoch = params.cur_epoch
+        params.best_train_loss = params.train_loss
+
+
+def run(rank, world_size, args):
+    """
+    Args:
+      rank:
+        It is a value between 0 and `world_size-1`, which is
+        passed automatically by `mp.spawn()` in :func:`main`.
+        The node with rank 0 is responsible for saving checkpoint.
+      world_size:
+        Number of GPUs for DDP training.
+      args:
+        The return value of get_parser().parse_args()
+    """
+    params = get_params()
+    params.update(vars(args))
+
+    fix_random_seed(params.seed)
+    if world_size > 1:
+        setup_dist(rank, world_size, params.master_port)
+
+    setup_logger(f"{params.exp_dir}/log/log-train")
+    logging.info("Training started")
+
+    if args.tensorboard and rank == 0:
+        tb_writer = SummaryWriter(log_dir=f"{params.exp_dir}/tensorboard")
+    else:
+        tb_writer = None
+
+    device = torch.device("cpu")
+    if torch.cuda.is_available():
+        device = torch.device("cuda", rank)
+    logging.info(f"Device: {device}")
+
+    sp = spm.SentencePieceProcessor()
+    sp.load(params.bpe_model)
+
+    # <blk> is defined in local/train_bpe_model.py
+    params.blank_id = sp.piece_to_id("<blk>")
+    params.vocab_size = sp.get_piece_size()
+
+    logging.info(params)
+
+    logging.info("About to create model")
+    model = get_transducer_model(params)
+
+    num_param = sum([p.numel() for p in model.parameters()])
+    logging.info(f"Number of model parameters: {num_param}")
+
+    checkpoints = load_checkpoint_if_available(params=params, model=model)
+
+    model.to(device)
+    if world_size > 1:
+        logging.info("Using DDP")
+        model = DDP(model, device_ids=[rank])
+    model.device = device
+
+    optimizer = Eve(model.parameters(), lr=params.initial_lr)
+
+    scheduler = Eden(optimizer, params.lr_batches, params.lr_epochs)
+
+    if checkpoints and "optimizer" in checkpoints:
+        logging.info("Loading optimizer state dict")
+        optimizer.load_state_dict(checkpoints["optimizer"])
+
+    if (
+        checkpoints
+        and "scheduler" in checkpoints
+        and checkpoints["scheduler"] is not None
+    ):
+        logging.info("Loading scheduler state dict")
+        scheduler.load_state_dict(checkpoints["scheduler"])
+
+    if params.print_diagnostics:
+        diagnostic = diagnostics.attach_diagnostics(model)
+
+    gigaspeech = GigaSpeechAsrDataModule(args)
+
+    train_cuts = gigaspeech.train_cuts()
+
+    if params.start_batch > 0 and checkpoints and "sampler" in checkpoints:
+        # We only load the sampler's state dict when it loads a checkpoint
+        # saved in the middle of an epoch
+        sampler_state_dict = checkpoints["sampler"]
+    else:
+        sampler_state_dict = None
+
+    train_dl = gigaspeech.train_dataloaders(
+        train_cuts, sampler_state_dict=sampler_state_dict
+    )
+
+    valid_cuts = gigaspeech.dev_cuts()
+    valid_dl = gigaspeech.valid_dataloaders(valid_cuts)
+
+    if not params.print_diagnostics:
+        scan_pessimistic_batches_for_oom(
+            model=model,
+            train_dl=train_dl,
+            optimizer=optimizer,
+            sp=sp,
+            params=params,
+        )
+
+    scaler = GradScaler(enabled=params.use_fp16)
+    if checkpoints and "grad_scaler" in checkpoints:
+        logging.info("Loading grad scaler state dict")
+        scaler.load_state_dict(checkpoints["grad_scaler"])
+
+    for epoch in range(params.start_epoch, params.num_epochs):
+        scheduler.step_epoch(epoch)
+        fix_random_seed(params.seed + epoch)
+        train_dl.sampler.set_epoch(epoch)
+
+        if tb_writer is not None:
+            tb_writer.add_scalar("train/epoch", epoch, params.batch_idx_train)
+
+        params.cur_epoch = epoch
+
+        train_one_epoch(
+            params=params,
+            model=model,
+            optimizer=optimizer,
+            scheduler=scheduler,
+            sp=sp,
+            train_dl=train_dl,
+            valid_dl=valid_dl,
+            scaler=scaler,
+            tb_writer=tb_writer,
+            world_size=world_size,
+            rank=rank,
+        )
+
+        if params.print_diagnostics:
+            diagnostic.print_diagnostics()
+            break
+
+        save_checkpoint(
+            params=params,
+            model=model,
+            optimizer=optimizer,
+            scheduler=scheduler,
+            sampler=train_dl.sampler,
+            scaler=scaler,
+            rank=rank,
+        )
+
+    logging.info("Done!")
+
+    if world_size > 1:
+        torch.distributed.barrier()
+        cleanup_dist()
+
+
+def scan_pessimistic_batches_for_oom(
+    model: nn.Module,
+    train_dl: torch.utils.data.DataLoader,
+    optimizer: torch.optim.Optimizer,
+    sp: spm.SentencePieceProcessor,
+    params: AttributeDict,
+):
+    from lhotse.dataset import find_pessimistic_batches
+
+    logging.info(
+        "Sanity check -- see if any of the batches in epoch 0 would cause OOM."
+    )
+    batches, crit_values = find_pessimistic_batches(train_dl.sampler)
+    for criterion, cuts in batches.items():
+        batch = train_dl.dataset[cuts]
+        try:
+            # warmup = 0.0 is so that the derivs for the pruned loss stay zero
+            # (i.e. are not remembered by the decaying-average in adam), because
+            # we want to avoid these params being subject to shrinkage in adam.
+            with torch.cuda.amp.autocast(enabled=params.use_fp16):
+                loss, _ = compute_loss(
+                    params=params,
+                    model=model,
+                    sp=sp,
+                    batch=batch,
+                    is_training=True,
+                    warmup=0.0,
+                )
+            loss.backward()
+            optimizer.step()
+            optimizer.zero_grad()
+        except RuntimeError as e:
+            if "CUDA out of memory" in str(e):
+                logging.error(
+                    "Your GPU ran out of memory with the current "
+                    "max_duration setting. We recommend decreasing "
+                    "max_duration and trying again.\n"
+                    f"Failing criterion: {criterion} "
+                    f"(={crit_values[criterion]}) ..."
+                )
+            raise
+
+
+def main():
+    parser = get_parser()
+    GigaSpeechAsrDataModule.add_arguments(parser)
+    args = parser.parse_args()
+    args.exp_dir = Path(args.exp_dir)
+
+    world_size = args.world_size
+    assert world_size >= 1
+    if world_size > 1:
+        mp.spawn(run, args=(world_size, args), nprocs=world_size, join=True)
+    else:
+        run(rank=0, world_size=1, args=args)
+
+
+torch.set_num_threads(1)
+torch.set_num_interop_threads(1)
+
+if __name__ == "__main__":
+    main()

egs/librispeech/ASR/RESULTS.md

@@ -133,7 +133,7 @@ results at:
 <https://huggingface.co/csukuangfj/icefall-asr-librispeech-pruned-transducer-stateless5-M-2022-05-13>
 
 
-### LibriSpeech BPE training results (Pruned Stateless Transducer 3)
+### LibriSpeech BPE training results (Pruned Stateless Transducer 3, 2022-04-29)
 
 [pruned_transducer_stateless3](./pruned_transducer_stateless3)
 Same as `Pruned Stateless Transducer 2` but using the XL subset from
@@ -286,6 +286,67 @@ for epoch in 27; do
 done
 ```
 
+### LibriSpeech BPE training results (Pruned Transducer 3, 2022-05-13)
+
+Same setup as [pruned_transducer_stateless3](./pruned_transducer_stateless3) (2022-04-29)
+but change `--giga-prob` from 0.8 to 0.9. Also use `repeat` on gigaspeech XL
+subset so that the gigaspeech dataloader never exhausts.
+
+|                                     | test-clean | test-other | comment                                                                       |
+|-------------------------------------|------------|------------|---------------------------------------------|
+| greedy search (max sym per frame 1) | 2.03       | 4.70       | --iter 1224000 --avg 14  --max-duration 600 |
+| modified beam search                | 2.00       | 4.63       | --iter 1224000 --avg 14  --max-duration 600 |
+| fast beam search                    | 2.10       | 4.68       | --iter 1224000 --avg 14 --max-duration 600 |
+
+The training commands are:
+
+```bash
+export CUDA_VISIBLE_DEVICES="0,1,2,3,4,5,6,7"
+
+./prepare.sh
+./prepare_giga_speech.sh
+
+./pruned_transducer_stateless3/train.py \
+  --world-size 8 \
+  --num-epochs 30 \
+  --start-epoch 0 \
+  --full-libri 1 \
+  --exp-dir pruned_transducer_stateless3/exp-0.9 \
+  --max-duration 300 \
+  --use-fp16 1 \
+  --lr-epochs 4 \
+  --num-workers 2 \
+  --giga-prob 0.9
+```
+
+The tensorboard log is available at
+<https://tensorboard.dev/experiment/HpocR7dKS9KCQkJeYxfXug/>
+
+Decoding commands:
+
+```bash
+for iter in 1224000; do
+  for avg in 14; do
+    for method in greedy_search modified_beam_search fast_beam_search ; do
+      ./pruned_transducer_stateless3/decode.py \
+        --iter $iter \
+        --avg $avg \
+        --exp-dir ./pruned_transducer_stateless3/exp-0.9/ \
+        --max-duration 600 \
+        --decoding-method $method \
+        --max-sym-per-frame 1 \
+        --beam 4 \
+        --max-contexts 32
+    done
+  done
+done
+```
+
+The pretrained models, training logs, decoding logs, and decoding results
+can be found at
+<https://huggingface.co/csukuangfj/icefall-asr-librispeech-pruned-transducer-stateless3-2022-05-13>
+
+
 ### LibriSpeech BPE training results (Pruned Transducer 2)
 
 [pruned_transducer_stateless2](./pruned_transducer_stateless2)

egs/librispeech/ASR/local/prepare_lang_bpe.py

@@ -145,7 +145,14 @@ def generate_lexicon(
     sp = spm.SentencePieceProcessor()
     sp.load(str(model_file))
 
-    words_pieces: List[List[str]] = sp.encode(words, out_type=str)
+    # Convert word to word piece IDs instead of word piece strings
+    # to avoid OOV tokens.
+    words_pieces_ids: List[List[int]] = sp.encode(words, out_type=int)
+
+    # Now convert word piece IDs back to word piece strings.
+    words_pieces: List[List[str]] = [
+        sp.id_to_piece(ids) for ids in words_pieces_ids
+    ]
 
     lexicon = []
     for word, pieces in zip(words, words_pieces):

egs/librispeech/ASR/local/validate_bpe_lexicon.py

@@ -0,0 +1,77 @@
+#!/usr/bin/env python3
+# Copyright    2022  Xiaomi Corp.        (authors: Fangjun Kuang)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+This script checks that there are no OOV tokens in the BPE-based lexicon.
+
+Usage example:
+
+    python3 ./local/validate_bpe_lexicon.py \
+            --lexicon /path/to/lexicon.txt \
+            --bpe-model /path/to/bpe.model
+"""
+
+import argparse
+from pathlib import Path
+from typing import List, Tuple
+
+import sentencepiece as spm
+
+from icefall.lexicon import read_lexicon
+
+# Map word to word pieces
+Lexicon = List[Tuple[str, List[str]]]
+
+
+def get_args():
+    parser = argparse.ArgumentParser()
+
+    parser.add_argument(
+        "--lexicon",
+        required=True,
+        type=Path,
+        help="Path to lexicon.txt",
+    )
+
+    parser.add_argument(
+        "--bpe-model",
+        required=True,
+        type=Path,
+        help="Path to bpe.model",
+    )
+
+    return parser.parse_args()
+
+
+def main():
+    args = get_args()
+    assert args.lexicon.is_file(), args.lexicon
+    assert args.bpe_model.is_file(), args.bpe_model
+
+    lexicon = read_lexicon(args.lexicon)
+
+    sp = spm.SentencePieceProcessor()
+    sp.load(str(args.bpe_model))
+
+    word_pieces = set(sp.id_to_piece(list(range(sp.vocab_size()))))
+    for word, pieces in lexicon:
+        for p in pieces:
+            if p not in word_pieces:
+                raise ValueError(f"The word {word} contains an OOV token {p}")
+
+
+if __name__ == "__main__":
+    main()

egs/librispeech/ASR/prepare.sh

@@ -184,13 +184,20 @@ if [ $stage -le 6 ] && [ $stop_stage -ge 6 ]; then
       done > $lang_dir/transcript_words.txt
     fi
 
-    ./local/train_bpe_model.py \
+    if [ ! -f $lang_dir/bpe.model ]; then
-      --lang-dir $lang_dir \
+      ./local/train_bpe_model.py \
-      --vocab-size $vocab_size \
+        --lang-dir $lang_dir \
-      --transcript $lang_dir/transcript_words.txt
+        --vocab-size $vocab_size \
+        --transcript $lang_dir/transcript_words.txt
+    fi
 
     if [ ! -f $lang_dir/L_disambig.pt ]; then
       ./local/prepare_lang_bpe.py --lang-dir $lang_dir
+
+      log "Validating $lang_dir/lexicon.txt"
+      ./local/validate_bpe_lexicon.py \
+        --lexicon $lang_dir/lexicon.txt \
+        --bpe-model $lang_dir/bpe.model
     fi
   done
 fi

egs/librispeech/ASR/pruned_transducer_stateless/export.py

@@ -116,8 +116,6 @@ def main():
     args = get_parser().parse_args()
     args.exp_dir = Path(args.exp_dir)
 
-    assert args.jit is False, "Support torchscript will be added later"
-
     params = get_params()
     params.update(vars(args))
 
@@ -159,6 +157,11 @@ def main():
     model.eval()
 
     if params.jit:
+        # We won't use the forward() method of the model in C++, so just ignore
+        # it here.
+        # Otherwise, one of its arguments is a ragged tensor and is not
+        # torch scriptabe.
+        model.__class__.forward = torch.jit.ignore(model.__class__.forward)
         logging.info("Using torch.jit.script")
         model = torch.jit.script(model)
         filename = params.exp_dir / "cpu_jit.pt"

egs/librispeech/ASR/pruned_transducer_stateless/pretrained.py

@@ -19,38 +19,38 @@ Usage:
 
 (1) greedy search
 ./pruned_transducer_stateless/pretrained.py \
-        --checkpoint ./pruned_transducer_stateless/exp/pretrained.pt \
+    --checkpoint ./pruned_transducer_stateless/exp/pretrained.pt \
-        --bpe-model ./data/lang_bpe_500/bpe.model \
+    --bpe-model ./data/lang_bpe_500/bpe.model \
-        --method greedy_search \
+    --method greedy_search \
-        /path/to/foo.wav \
+    /path/to/foo.wav \
-        /path/to/bar.wav \
+    /path/to/bar.wav
 
 (2) beam search
 ./pruned_transducer_stateless/pretrained.py \
-        --checkpoint ./pruned_transducer_stateless/exp/pretrained.pt \
+    --checkpoint ./pruned_transducer_stateless/exp/pretrained.pt \
-        --bpe-model ./data/lang_bpe_500/bpe.model \
+    --bpe-model ./data/lang_bpe_500/bpe.model \
-        --method beam_search \
+    --method beam_search \
-        --beam-size 4 \
+    --beam-size 4 \
-        /path/to/foo.wav \
+    /path/to/foo.wav \
-        /path/to/bar.wav \
+    /path/to/bar.wav
 
 (3) modified beam search
 ./pruned_transducer_stateless/pretrained.py \
-        --checkpoint ./pruned_transducer_stateless/exp/pretrained.pt \
+    --checkpoint ./pruned_transducer_stateless/exp/pretrained.pt \
-        --bpe-model ./data/lang_bpe_500/bpe.model \
+    --bpe-model ./data/lang_bpe_500/bpe.model \
-        --method modified_beam_search \
+    --method modified_beam_search \
-        --beam-size 4 \
+    --beam-size 4 \
-        /path/to/foo.wav \
+    /path/to/foo.wav \
-        /path/to/bar.wav \
+    /path/to/bar.wav
 
 (4) fast beam search
 ./pruned_transducer_stateless/pretrained.py \
-        --checkpoint ./pruned_transducer_stateless/exp/pretrained.pt \
+    --checkpoint ./pruned_transducer_stateless/exp/pretrained.pt \
-        --bpe-model ./data/lang_bpe_500/bpe.model \
+    --bpe-model ./data/lang_bpe_500/bpe.model \
-        --method fast_beam_search \
+    --method fast_beam_search \
-        --beam-size 4 \
+    --beam-size 4 \
-        /path/to/foo.wav \
+    /path/to/foo.wav \
-        /path/to/bar.wav \
+    /path/to/bar.wav
 
 You can also use `./pruned_transducer_stateless/exp/epoch-xx.pt`.
 
@@ -233,6 +233,9 @@ def main():
     logging.info("Creating model")
     model = get_transducer_model(params)
 
+    num_param = sum([p.numel() for p in model.parameters()])
+    logging.info(f"Number of model parameters: {num_param}")
+
     checkpoint = torch.load(args.checkpoint, map_location="cpu")
     model.load_state_dict(checkpoint["model"], strict=False)
     model.to(device)

egs/librispeech/ASR/pruned_transducer_stateless/test_decoder.py

@@ -29,6 +29,7 @@ from decoder import Decoder
 def test_decoder():
     vocab_size = 3
     blank_id = 0
+    unk_id = 2
     embedding_dim = 128
     context_size = 4
 
@@ -36,6 +37,7 @@ def test_decoder():
         vocab_size=vocab_size,
         embedding_dim=embedding_dim,
         blank_id=blank_id,
+        unk_id=unk_id,
         context_size=context_size,
     )
     N = 100

egs/librispeech/ASR/pruned_transducer_stateless/test_model.py

@@ -0,0 +1,50 @@
+#!/usr/bin/env python3
+# Copyright    2022  Xiaomi Corp.        (authors: Fangjun Kuang)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+"""
+To run this file, do:
+
+    cd icefall/egs/librispeech/ASR
+    python ./pruned_transducer_stateless/test_model.py
+"""
+
+import torch
+from train import get_params, get_transducer_model
+
+
+def test_model():
+    params = get_params()
+    params.vocab_size = 500
+    params.blank_id = 0
+    params.context_size = 2
+    params.unk_id = 2
+
+    model = get_transducer_model(params)
+
+    num_param = sum([p.numel() for p in model.parameters()])
+    print(f"Number of model parameters: {num_param}")
+    model.__class__.forward = torch.jit.ignore(model.__class__.forward)
+    torch.jit.script(model)
+
+
+def main():
+    test_model()
+
+
+if __name__ == "__main__":
+    main()

egs/librispeech/ASR/pruned_transducer_stateless2/conformer.py

@@ -112,10 +112,13 @@ class Conformer(EncoderInterface):
         x, pos_emb = self.encoder_pos(x)
         x = x.permute(1, 0, 2)  # (N, T, C) -> (T, N, C)
 
-        with warnings.catch_warnings():
+        # Caution: We assume the subsampling factor is 4!
-            warnings.simplefilter("ignore")
+
-            # Caution: We assume the subsampling factor is 4!
+        #  lengths = ((x_lens - 1) // 2 - 1) // 2 # issue an warning
-            lengths = ((x_lens - 1) // 2 - 1) // 2
+        #
+        # Note: rounding_mode in torch.div() is available only in torch >= 1.8.0
+        lengths = (((x_lens - 1) >> 1) - 1) >> 1
+
         assert x.size(0) == lengths.max().item()
         mask = make_pad_mask(lengths)
 

egs/librispeech/ASR/pruned_transducer_stateless2/export.py

@@ -51,7 +51,11 @@ import sentencepiece as spm
 import torch
 from train import get_params, get_transducer_model
 
-from icefall.checkpoint import average_checkpoints, load_checkpoint
+from icefall.checkpoint import (
+    average_checkpoints,
+    find_checkpoints,
+    load_checkpoint,
+)
 from icefall.utils import str2bool
 
 
@@ -64,8 +68,19 @@ def get_parser():
         "--epoch",
         type=int,
         default=28,
-        help="It specifies the checkpoint to use for decoding."
+        help="""It specifies the checkpoint to use for averaging.
-        "Note: Epoch counts from 0.",
+        Note: Epoch counts from 0.
+        You can specify --avg to use more checkpoints for model averaging.""",
+    )
+
+    parser.add_argument(
+        "--iter",
+        type=int,
+        default=0,
+        help="""If positive, --epoch is ignored and it
+        will use the checkpoint exp_dir/checkpoint-iter.pt.
+        You can specify --avg to use more checkpoints for model averaging.
+        """,
     )
 
     parser.add_argument(
@@ -74,7 +89,7 @@ def get_parser():
         default=15,
         help="Number of checkpoints to average. Automatically select "
         "consecutive checkpoints before the checkpoint specified by "
-        "'--epoch'. ",
+        "'--epoch' and '--iter'",
     )
 
     parser.add_argument(
@@ -116,8 +131,6 @@ def main():
     args = get_parser().parse_args()
     args.exp_dir = Path(args.exp_dir)
 
-    assert args.jit is False, "Support torchscript will be added later"
-
     params = get_params()
     params.update(vars(args))
 
@@ -141,7 +154,24 @@ def main():
 
     model.to(device)
 
-    if params.avg == 1:
+    if params.iter > 0:
+        filenames = find_checkpoints(params.exp_dir, iteration=-params.iter)[
+            : params.avg
+        ]
+        if len(filenames) == 0:
+            raise ValueError(
+                f"No checkpoints found for"
+                f" --iter {params.iter}, --avg {params.avg}"
+            )
+        elif len(filenames) < params.avg:
+            raise ValueError(
+                f"Not enough checkpoints ({len(filenames)}) found for"
+                f" --iter {params.iter}, --avg {params.avg}"
+            )
+        logging.info(f"averaging {filenames}")
+        model.to(device)
+        model.load_state_dict(average_checkpoints(filenames, device=device))
+    elif params.avg == 1:
         load_checkpoint(f"{params.exp_dir}/epoch-{params.epoch}.pt", model)
     else:
         start = params.epoch - params.avg + 1
@@ -159,6 +189,11 @@ def main():
     model.eval()
 
     if params.jit:
+        # We won't use the forward() method of the model in C++, so just ignore
+        # it here.
+        # Otherwise, one of its arguments is a ragged tensor and is not
+        # torch scriptabe.
+        model.__class__.forward = torch.jit.ignore(model.__class__.forward)
         logging.info("Using torch.jit.script")
         model = torch.jit.script(model)
         filename = params.exp_dir / "cpu_jit.pt"

egs/librispeech/ASR/pruned_transducer_stateless2/pretrained.py

@@ -19,20 +19,38 @@ Usage:
 
 (1) greedy search
 ./pruned_transducer_stateless2/pretrained.py \
-        --checkpoint ./pruned_transducer_stateless2/exp/pretrained.pt \
+    --checkpoint ./pruned_transducer_stateless2/exp/pretrained.pt \
-        --bpe-model ./data/lang_bpe_500/bpe.model \
+    --bpe-model ./data/lang_bpe_500/bpe.model \
-        --method greedy_search \
+    --method greedy_search \
-        /path/to/foo.wav \
+    /path/to/foo.wav \
-        /path/to/bar.wav \
+    /path/to/bar.wav
 
-(1) beam search
+(2) beam search
 ./pruned_transducer_stateless2/pretrained.py \
-        --checkpoint ./pruned_transducer_stateless2/exp/pretrained.pt \
+    --checkpoint ./pruned_transducer_stateless2/exp/pretrained.pt \
-        --bpe-model ./data/lang_bpe_500/bpe.model \
+    --bpe-model ./data/lang_bpe_500/bpe.model \
-        --method beam_search \
+    --method beam_search \
-        --beam-size 4 \
+    --beam-size 4 \
-        /path/to/foo.wav \
+    /path/to/foo.wav \
-        /path/to/bar.wav \
+    /path/to/bar.wav
+
+(3) modified beam search
+./pruned_transducer_stateless2/pretrained.py \
+    --checkpoint ./pruned_transducer_stateless2/exp/pretrained.pt \
+    --bpe-model ./data/lang_bpe_500/bpe.model \
+    --method modified_beam_search \
+    --beam-size 4 \
+    /path/to/foo.wav \
+    /path/to/bar.wav
+
+(4) fast beam search
+./pruned_transducer_stateless2/pretrained.py \
+    --checkpoint ./pruned_transducer_stateless2/exp/pretrained.pt \
+    --bpe-model ./data/lang_bpe_500/bpe.model \
+    --method fast_beam_search \
+    --beam-size 4 \
+    /path/to/foo.wav \
+    /path/to/bar.wav
 
 You can also use `./pruned_transducer_stateless2/exp/epoch-xx.pt`.
 
@@ -79,9 +97,7 @@ def get_parser():
     parser.add_argument(
         "--bpe-model",
         type=str,
-        help="""Path to bpe.model.
+        help="""Path to bpe.model.""",
-        Used only when method is ctc-decoding.
-        """,
     )
 
     parser.add_argument(
@@ -117,7 +133,33 @@ def get_parser():
         "--beam-size",
         type=int,
         default=4,
-        help="Used only when --method is beam_search and modified_beam_search",
+        help="""An integer indicating how many candidates we will keep for each
+        frame. Used only when --method is beam_search or
+        modified_beam_search.""",
+    )
+
+    parser.add_argument(
+        "--beam",
+        type=float,
+        default=4,
+        help="""A floating point value to calculate the cutoff score during beam
+        search (i.e., `cutoff = max-score - beam`), which is the same as the
+        `beam` in Kaldi.
+        Used only when --method is fast_beam_search""",
+    )
+
+    parser.add_argument(
+        "--max-contexts",
+        type=int,
+        default=4,
+        help="""Used only when --method is fast_beam_search""",
+    )
+
+    parser.add_argument(
+        "--max-states",
+        type=int,
+        default=8,
+        help="""Used only when --method is fast_beam_search""",
     )
 
     parser.add_argument(
@@ -244,9 +286,9 @@ def main():
             decoding_graph=decoding_graph,
             encoder_out=encoder_out,
             encoder_out_lens=encoder_out_lens,
-            beam=8.0,
+            beam=params.beam,
-            max_contexts=32,
+            max_contexts=params.max_contexts,
-            max_states=8,
+            max_states=params.max_states,
         )
         for hyp in sp.decode(hyp_tokens):
             hyps.append(hyp.split())
@@ -254,6 +296,7 @@ def main():
         hyp_tokens = modified_beam_search(
             model=model,
             encoder_out=encoder_out,
+            encoder_out_lens=encoder_out_lens,
             beam=params.beam_size,
         )
 
@@ -263,6 +306,7 @@ def main():
         hyp_tokens = greedy_search_batch(
             model=model,
             encoder_out=encoder_out,
+            encoder_out_lens=encoder_out_lens,
         )
         for hyp in sp.decode(hyp_tokens):
             hyps.append(hyp.split())

egs/librispeech/ASR/pruned_transducer_stateless2/scaling.py

@@ -212,7 +212,10 @@ class ScaledLinear(nn.Linear):
         return self.weight * self.weight_scale.exp()
 
     def get_bias(self):
-        return None if self.bias is None else self.bias * self.bias_scale.exp()
+        if self.bias is None or self.bias_scale is None:
+            return None
+
+        return self.bias * self.bias_scale.exp()
 
     def forward(self, input: Tensor) -> Tensor:
         return torch.nn.functional.linear(
@@ -255,7 +258,11 @@ class ScaledConv1d(nn.Conv1d):
         return self.weight * self.weight_scale.exp()
 
     def get_bias(self):
-        return None if self.bias is None else self.bias * self.bias_scale.exp()
+        bias = self.bias
+        bias_scale = self.bias_scale
+        if bias is None or bias_scale is None:
+            return None
+        return bias * bias_scale.exp()
 
     def forward(self, input: Tensor) -> Tensor:
         F = torch.nn.functional
@@ -269,7 +276,7 @@ class ScaledConv1d(nn.Conv1d):
                 self.get_weight(),
                 self.get_bias(),
                 self.stride,
-                _single(0),
+                (0,),
                 self.dilation,
                 self.groups,
             )
@@ -319,7 +326,12 @@ class ScaledConv2d(nn.Conv2d):
         return self.weight * self.weight_scale.exp()
 
     def get_bias(self):
-        return None if self.bias is None else self.bias * self.bias_scale.exp()
+        # see https://github.com/pytorch/pytorch/issues/24135
+        bias = self.bias
+        bias_scale = self.bias_scale
+        if bias is None or bias_scale is None:
+            return None
+        return bias * bias_scale.exp()
 
     def _conv_forward(self, input, weight):
         F = torch.nn.functional
@@ -333,7 +345,7 @@ class ScaledConv2d(nn.Conv2d):
                 weight,
                 self.get_bias(),
                 self.stride,
-                _pair(0),
+                (0, 0),
                 self.dilation,
                 self.groups,
             )
@@ -398,6 +410,9 @@ class ActivationBalancer(torch.nn.Module):
         self.max_abs = max_abs
 
     def forward(self, x: Tensor) -> Tensor:
+        if torch.jit.is_scripting():
+            return x
+
         return ActivationBalancerFunction.apply(
             x,
             self.channel_dim,
@@ -444,6 +459,8 @@ class DoubleSwish(torch.nn.Module):
         """Return double-swish activation function which is an approximation to Swish(Swish(x)),
         that we approximate closely with x * sigmoid(x-1).
         """
+        if torch.jit.is_scripting():
+            return x * torch.sigmoid(x - 1.0)
         return DoubleSwishFunction.apply(x)
 
 

egs/librispeech/ASR/pruned_transducer_stateless2/test_model.py

@@ -23,6 +23,7 @@ To run this file, do:
     python ./pruned_transducer_stateless2/test_model.py
 """
 
+import torch
 from train import get_params, get_transducer_model
 
 
@@ -31,9 +32,14 @@ def test_model():
     params.vocab_size = 500
     params.blank_id = 0
     params.context_size = 2
+    params.unk_id = 2
+
     model = get_transducer_model(params)
+
     num_param = sum([p.numel() for p in model.parameters()])
     print(f"Number of model parameters: {num_param}")
+    model.__class__.forward = torch.jit.ignore(model.__class__.forward)
+    torch.jit.script(model)
 
 
 def main():

egs/librispeech/ASR/pruned_transducer_stateless2/train.py

@@ -695,7 +695,7 @@ def train_one_epoch(
             display_and_save_batch(batch, params=params, sp=sp)
             raise
 
-        if params.print_diagnostics and batch_idx == 5:
+        if params.print_diagnostics and batch_idx == 30:
             return
 
         if (
@@ -839,10 +839,7 @@ def run(rank, world_size, args):
         scheduler.load_state_dict(checkpoints["scheduler"])
 
     if params.print_diagnostics:
-        opts = diagnostics.TensorDiagnosticOptions(
+        diagnostic = diagnostics.attach_diagnostics(model)
-            2 ** 22
-        )  # allow 4 megabytes per sub-module
-        diagnostic = diagnostics.attach_diagnostics(model, opts)
 
     librispeech = LibriSpeechAsrDataModule(args)
 

egs/librispeech/ASR/pruned_transducer_stateless3/export.py

@@ -52,7 +52,11 @@ import sentencepiece as spm
 import torch
 from train import get_params, get_transducer_model
 
-from icefall.checkpoint import average_checkpoints, load_checkpoint
+from icefall.checkpoint import (
+    average_checkpoints,
+    find_checkpoints,
+    load_checkpoint,
+)
 from icefall.utils import str2bool
 
 
@@ -65,8 +69,19 @@ def get_parser():
         "--epoch",
         type=int,
         default=28,
-        help="It specifies the checkpoint to use for decoding."
+        help="""It specifies the checkpoint to use for averaging.
-        "Note: Epoch counts from 0.",
+        Note: Epoch counts from 0.
+        You can specify --avg to use more checkpoints for model averaging.""",
+    )
+
+    parser.add_argument(
+        "--iter",
+        type=int,
+        default=0,
+        help="""If positive, --epoch is ignored and it
+        will use the checkpoint exp_dir/checkpoint-iter.pt.
+        You can specify --avg to use more checkpoints for model averaging.
+        """,
     )
 
     parser.add_argument(
@@ -75,7 +90,7 @@ def get_parser():
         default=15,
         help="Number of checkpoints to average. Automatically select "
         "consecutive checkpoints before the checkpoint specified by "
-        "'--epoch'. ",
+        "'--epoch' and '--iter'",
     )
 
     parser.add_argument(
@@ -117,8 +132,6 @@ def main():
     args = get_parser().parse_args()
     args.exp_dir = Path(args.exp_dir)
 
-    assert args.jit is False, "Support torchscript will be added later"
-
     params = get_params()
     params.update(vars(args))
 
@@ -142,7 +155,24 @@ def main():
 
     model.to(device)
 
-    if params.avg == 1:
+    if params.iter > 0:
+        filenames = find_checkpoints(params.exp_dir, iteration=-params.iter)[
+            : params.avg
+        ]
+        if len(filenames) == 0:
+            raise ValueError(
+                f"No checkpoints found for"
+                f" --iter {params.iter}, --avg {params.avg}"
+            )
+        elif len(filenames) < params.avg:
+            raise ValueError(
+                f"Not enough checkpoints ({len(filenames)}) found for"
+                f" --iter {params.iter}, --avg {params.avg}"
+            )
+        logging.info(f"averaging {filenames}")
+        model.to(device)
+        model.load_state_dict(average_checkpoints(filenames, device=device))
+    elif params.avg == 1:
         load_checkpoint(f"{params.exp_dir}/epoch-{params.epoch}.pt", model)
     else:
         start = params.epoch - params.avg + 1
@@ -160,6 +190,11 @@ def main():
     model.eval()
 
     if params.jit:
+        # We won't use the forward() method of the model in C++, so just ignore
+        # it here.
+        # Otherwise, one of its arguments is a ragged tensor and is not
+        # torch scriptabe.
+        model.__class__.forward = torch.jit.ignore(model.__class__.forward)
         logging.info("Using torch.jit.script")
         model = torch.jit.script(model)
         filename = params.exp_dir / "cpu_jit.pt"

egs/librispeech/ASR/pruned_transducer_stateless3/pretrained.py

@@ -19,20 +19,38 @@ Usage:
 
 (1) greedy search
 ./pruned_transducer_stateless3/pretrained.py \
-        --checkpoint ./pruned_transducer_stateless3/exp/pretrained.pt \
+    --checkpoint ./pruned_transducer_stateless3/exp/pretrained.pt \
-        --bpe-model ./data/lang_bpe_500/bpe.model \
+    --bpe-model ./data/lang_bpe_500/bpe.model \
-        --method greedy_search \
+    --method greedy_search \
-        /path/to/foo.wav \
+    /path/to/foo.wav \
-        /path/to/bar.wav \
+    /path/to/bar.wav
 
-(1) beam search
+(2) beam search
 ./pruned_transducer_stateless3/pretrained.py \
-        --checkpoint ./pruned_transducer_stateless3/exp/pretrained.pt \
+    --checkpoint ./pruned_transducer_stateless3/exp/pretrained.pt \
-        --bpe-model ./data/lang_bpe_500/bpe.model \
+    --bpe-model ./data/lang_bpe_500/bpe.model \
-        --method beam_search \
+    --method beam_search \
-        --beam-size 4 \
+    --beam-size 4 \
-        /path/to/foo.wav \
+    /path/to/foo.wav \
-        /path/to/bar.wav \
+    /path/to/bar.wav
+
+(3) modified beam search
+./pruned_transducer_stateless3/pretrained.py \
+    --checkpoint ./pruned_transducer_stateless3/exp/pretrained.pt \
+    --bpe-model ./data/lang_bpe_500/bpe.model \
+    --method modified_beam_search \
+    --beam-size 4 \
+    /path/to/foo.wav \
+    /path/to/bar.wav
+
+(4) fast beam search
+./pruned_transducer_stateless3/pretrained.py \
+    --checkpoint ./pruned_transducer_stateless3/exp/pretrained.pt \
+    --bpe-model ./data/lang_bpe_500/bpe.model \
+    --method fast_beam_search \
+    --beam-size 4 \
+    /path/to/foo.wav \
+    /path/to/bar.wav
 
 You can also use `./pruned_transducer_stateless3/exp/epoch-xx.pt`.
 
@@ -79,9 +97,7 @@ def get_parser():
     parser.add_argument(
         "--bpe-model",
         type=str,
-        help="""Path to bpe.model.
+        help="""Path to bpe.model.""",
-        Used only when method is ctc-decoding.
-        """,
     )
 
     parser.add_argument(
@@ -117,7 +133,33 @@ def get_parser():
         "--beam-size",
         type=int,
         default=4,
-        help="Used only when --method is beam_search and modified_beam_search",
+        help="""An integer indicating how many candidates we will keep for each
+        frame. Used only when --method is beam_search or
+        modified_beam_search.""",
+    )
+
+    parser.add_argument(
+        "--beam",
+        type=float,
+        default=4,
+        help="""A floating point value to calculate the cutoff score during beam
+        search (i.e., `cutoff = max-score - beam`), which is the same as the
+        `beam` in Kaldi.
+        Used only when --method is fast_beam_search""",
+    )
+
+    parser.add_argument(
+        "--max-contexts",
+        type=int,
+        default=4,
+        help="""Used only when --method is fast_beam_search""",
+    )
+
+    parser.add_argument(
+        "--max-states",
+        type=int,
+        default=8,
+        help="""Used only when --method is fast_beam_search""",
     )
 
     parser.add_argument(
@@ -244,9 +286,9 @@ def main():
             decoding_graph=decoding_graph,
             encoder_out=encoder_out,
             encoder_out_lens=encoder_out_lens,
-            beam=8.0,
+            beam=params.beam,
-            max_contexts=32,
+            max_contexts=params.max_contexts,
-            max_states=8,
+            max_states=params.max_states,
         )
         for hyp in sp.decode(hyp_tokens):
             hyps.append(hyp.split())
@@ -254,6 +296,7 @@ def main():
         hyp_tokens = modified_beam_search(
             model=model,
             encoder_out=encoder_out,
+            encoder_out_lens=encoder_out_lens,
             beam=params.beam_size,
         )
 
@@ -263,6 +306,7 @@ def main():
         hyp_tokens = greedy_search_batch(
             model=model,
             encoder_out=encoder_out,
+            encoder_out_lens=encoder_out_lens,
         )
         for hyp in sp.decode(hyp_tokens):
             hyps.append(hyp.split())

egs/librispeech/ASR/pruned_transducer_stateless3/test_model.py

@@ -0,0 +1,50 @@
+#!/usr/bin/env python3
+# Copyright    2022  Xiaomi Corp.        (authors: Fangjun Kuang)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+"""
+To run this file, do:
+
+    cd icefall/egs/librispeech/ASR
+    python ./pruned_transducer_stateless3/test_model.py
+"""
+
+import torch
+from train import get_params, get_transducer_model
+
+
+def test_model():
+    params = get_params()
+    params.vocab_size = 500
+    params.blank_id = 0
+    params.context_size = 2
+    params.unk_id = 2
+
+    model = get_transducer_model(params)
+
+    num_param = sum([p.numel() for p in model.parameters()])
+    print(f"Number of model parameters: {num_param}")
+    model.__class__.forward = torch.jit.ignore(model.__class__.forward)
+    torch.jit.script(model)
+
+
+def main():
+    test_model()
+
+
+if __name__ == "__main__":
+    main()

egs/librispeech/ASR/pruned_transducer_stateless3/test_scaling.py

@@ -0,0 +1,69 @@
+#!/usr/bin/env python3
+# Copyright    2022  Xiaomi Corp.        (authors: Fangjun Kuang)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+"""
+To run this file, do:
+
+    cd icefall/egs/librispeech/ASR
+    python ./pruned_transducer_stateless3/test_scaling.py
+"""
+
+import torch
+from scaling import ActivationBalancer, ScaledConv1d, ScaledConv2d
+
+
+def test_scaled_conv1d():
+    for bias in [True, False]:
+        conv1d = ScaledConv1d(
+            3,
+            6,
+            kernel_size=1,
+            stride=1,
+            padding=0,
+            bias=bias,
+        )
+        torch.jit.script(conv1d)
+
+
+def test_scaled_conv2d():
+    for bias in [True, False]:
+        conv2d = ScaledConv2d(
+            in_channels=1,
+            out_channels=3,
+            kernel_size=3,
+            padding=1,
+            bias=bias,
+        )
+        torch.jit.script(conv2d)
+
+
+def test_activation_balancer():
+    act = ActivationBalancer(
+        channel_dim=1, max_abs=10.0, min_positive=0.05, max_positive=1.0
+    )
+    torch.jit.script(act)
+
+
+def main():
+    test_scaled_conv1d()
+    test_scaled_conv2d()
+    test_activation_balancer()
+
+
+if __name__ == "__main__":
+    main()

egs/librispeech/ASR/pruned_transducer_stateless3/train.py

@@ -767,7 +767,7 @@ def train_one_epoch(
         scaler.update()
         optimizer.zero_grad()
 
-        if params.print_diagnostics and batch_idx == 5:
+        if params.print_diagnostics and batch_idx == 30:
             return
 
         if (
@@ -938,10 +938,7 @@ def run(rank, world_size, args):
         scheduler.load_state_dict(checkpoints["scheduler"])
 
     if params.print_diagnostics:
-        opts = diagnostics.TensorDiagnosticOptions(
+        diagnostic = diagnostics.attach_diagnostics(model)
-            2 ** 22
-        )  # allow 4 megabytes per sub-module
-        diagnostic = diagnostics.attach_diagnostics(model, opts)
 
     librispeech = LibriSpeech(manifest_dir=args.manifest_dir)
 
@@ -968,6 +965,7 @@ def run(rank, world_size, args):
         train_giga_cuts = gigaspeech.train_S_cuts()
 
     train_giga_cuts = filter_short_and_long_utterances(train_giga_cuts)
+    train_giga_cuts = train_giga_cuts.repeat(times=None)
 
     if args.enable_musan:
         cuts_musan = load_manifest(

egs/librispeech/ASR/pruned_transducer_stateless4/test_model.py

@@ -0,0 +1,50 @@
+#!/usr/bin/env python3
+# Copyright    2022  Xiaomi Corp.        (authors: Fangjun Kuang)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+"""
+To run this file, do:
+
+    cd icefall/egs/librispeech/ASR
+    python ./pruned_transducer_stateless4/test_model.py
+"""
+
+import torch
+from train import get_params, get_transducer_model
+
+
+def test_model():
+    params = get_params()
+    params.vocab_size = 500
+    params.blank_id = 0
+    params.context_size = 2
+    params.unk_id = 2
+
+    model = get_transducer_model(params)
+
+    num_param = sum([p.numel() for p in model.parameters()])
+    print(f"Number of model parameters: {num_param}")
+    model.__class__.forward = torch.jit.ignore(model.__class__.forward)
+    torch.jit.script(model)
+
+
+def main():
+    test_model()
+
+
+if __name__ == "__main__":
+    main()

egs/librispeech/ASR/pruned_transducer_stateless4/train.py

@@ -724,7 +724,7 @@ def train_one_epoch(
         scaler.update()
         optimizer.zero_grad()
 
-        if params.print_diagnostics and batch_idx == 5:
+        if params.print_diagnostics and batch_idx == 30:
             return
 
         if (
@@ -888,10 +888,7 @@ def run(rank, world_size, args):
         scheduler.load_state_dict(checkpoints["scheduler"])
 
     if params.print_diagnostics:
-        opts = diagnostics.TensorDiagnosticOptions(
+        diagnostic = diagnostics.attach_diagnostics(model)
-            2 ** 22
-        )  # allow 4 megabytes per sub-module
-        diagnostic = diagnostics.attach_diagnostics(model, opts)
 
     librispeech = LibriSpeechAsrDataModule(args)
 

egs/librispeech/ASR/transducer_lstm/encoder.py

@@ -94,7 +94,7 @@ class LstmEncoder(EncoderInterface):
         )
 
         if False:
-            # It is commented out as DPP complains that not all parameters are
+            # It is commented out as DDP complains that not all parameters are
             # used. Need more checks later for the reason.
             #
             # Caution: We assume the dataloader returns utterances with
@@ -107,7 +107,7 @@ class LstmEncoder(EncoderInterface):
             )
 
             packed_rnn_out, _ = self.rnn(packed_x)
-            rnn_out, _ = pad_packed_sequence(packed_x, batch_first=True)
+            rnn_out, _ = pad_packed_sequence(packed_rnn_out, batch_first=True)
         else:
             rnn_out, _ = self.rnn(x)
 

egs/librispeech/ASR/transducer_lstm/model.py

@@ -97,8 +97,7 @@ class Transducer(nn.Module):
         y_lens = row_splits[1:] - row_splits[:-1]
 
         blank_id = self.decoder.blank_id
-        sos_id = self.decoder.sos_id
+        sos_y = add_sos(y, sos_id=blank_id)
-        sos_y = add_sos(y, sos_id=sos_id)
 
         sos_y_padded = sos_y.pad(mode="constant", padding_value=blank_id)
         sos_y_padded = sos_y_padded.to(torch.int64)

egs/librispeech/ASR/transducer_stateless/conformer.py

@@ -109,10 +109,12 @@ class Conformer(Transformer):
         x, pos_emb = self.encoder_pos(x)
         x = x.permute(1, 0, 2)  # (N, T, C) -> (T, N, C)
 
-        with warnings.catch_warnings():
+        # Caution: We assume the subsampling factor is 4!
-            warnings.simplefilter("ignore")
+
-            # Caution: We assume the subsampling factor is 4!
+        #  lengths = ((x_lens - 1) // 2 - 1) // 2 # issue an warning
-            lengths = ((x_lens - 1) // 2 - 1) // 2
+        #
+        # Note: rounding_mode in torch.div() is available only in torch >= 1.8.0
+        lengths = (((x_lens - 1) >> 1) - 1) >> 1
 
         assert x.size(0) == lengths.max().item()
         mask = make_pad_mask(lengths)

egs/librispeech/ASR/transducer_stateless/export.py

@@ -183,8 +183,6 @@ def main():
     args = get_parser().parse_args()
     args.exp_dir = Path(args.exp_dir)
 
-    assert args.jit is False, "Support torchscript will be added later"
-
     params = get_params()
     params.update(vars(args))
 
@@ -226,6 +224,11 @@ def main():
     model.eval()
 
     if params.jit:
+        # We won't use the forward() method of the model in C++, so just ignore
+        # it here.
+        # Otherwise, one of its arguments is a ragged tensor and is not
+        # torch scriptabe.
+        model.__class__.forward = torch.jit.ignore(model.__class__.forward)
         logging.info("Using torch.jit.script")
         model = torch.jit.script(model)
         filename = params.exp_dir / "cpu_jit.pt"

egs/librispeech/ASR/transducer_stateless/joiner.py

@@ -14,6 +14,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from typing import List
+
 import torch
 import torch.nn as nn
 
@@ -55,8 +57,8 @@ class Joiner(nn.Module):
 
         N = encoder_out.size(0)
 
-        encoder_out_len = encoder_out_len.tolist()
+        encoder_out_len: List[int] = encoder_out_len.tolist()
-        decoder_out_len = decoder_out_len.tolist()
+        decoder_out_len: List[int] = decoder_out_len.tolist()
 
         encoder_out_list = [
             encoder_out[i, : encoder_out_len[i], :] for i in range(N)

egs/librispeech/ASR/transducer_stateless/test_model.py

@@ -0,0 +1,49 @@
+#!/usr/bin/env python3
+# Copyright    2022  Xiaomi Corp.        (authors: Fangjun Kuang)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+"""
+To run this file, do:
+
+    cd icefall/egs/librispeech/ASR
+    python ./transducer_stateless/test_model.py
+"""
+
+import torch
+from train import get_params, get_transducer_model
+
+
+def test_model():
+    params = get_params()
+    params.vocab_size = 500
+    params.blank_id = 0
+    params.context_size = 2
+
+    model = get_transducer_model(params)
+
+    num_param = sum([p.numel() for p in model.parameters()])
+    print(f"Number of model parameters: {num_param}")
+    model.__class__.forward = torch.jit.ignore(model.__class__.forward)
+    torch.jit.script(model)
+
+
+def main():
+    test_model()
+
+
+if __name__ == "__main__":
+    main()

egs/librispeech/ASR/transducer_stateless/train.py

@@ -523,7 +523,7 @@ def train_one_epoch(
         loss.backward()
         clip_grad_norm_(model.parameters(), 5.0, 2.0)
         optimizer.step()
-        if params.print_diagnostics and batch_idx == 5:
+        if params.print_diagnostics and batch_idx == 30:
             return
 
         if batch_idx % params.log_interval == 0:
@@ -635,10 +635,7 @@ def run(rank, world_size, args):
     librispeech = LibriSpeechAsrDataModule(args)
 
     if params.print_diagnostics:
-        opts = diagnostics.TensorDiagnosticOptions(
+        diagnostic = diagnostics.attach_diagnostics(model)
-            2 ** 22
-        )  # allow 4 megabytes per sub-module
-        diagnostic = diagnostics.attach_diagnostics(model, opts)
 
     train_cuts = librispeech.train_clean_100_cuts()
     if params.full_libri:

egs/librispeech/ASR/transducer_stateless2/export.py

@@ -115,8 +115,6 @@ def main():
     args = get_parser().parse_args()
     args.exp_dir = Path(args.exp_dir)
 
-    assert args.jit is False, "Support torchscript will be added later"
-
     params = get_params()
     params.update(vars(args))
 
@@ -158,6 +156,11 @@ def main():
     model.eval()
 
     if params.jit:
+        # We won't use the forward() method of the model in C++, so just ignore
+        # it here.
+        # Otherwise, one of its arguments is a ragged tensor and is not
+        # torch scriptabe.
+        model.__class__.forward = torch.jit.ignore(model.__class__.forward)
         logging.info("Using torch.jit.script")
         model = torch.jit.script(model)
         filename = params.exp_dir / "cpu_jit.pt"

egs/librispeech/ASR/transducer_stateless2/joiner.py

@@ -14,6 +14,8 @@
 # See the License for the specific language governing permissions and
 # limitations under the License.
 
+from typing import Optional
+
 import torch
 import torch.nn as nn
 
@@ -30,7 +32,8 @@ class Joiner(nn.Module):
         self,
         encoder_out: torch.Tensor,
         decoder_out: torch.Tensor,
-        *unused,
+        unused_encoder_out_len: Optional[torch.Tensor] = None,
+        unused_decoder_out_len: Optional[torch.Tensor] = None,
     ) -> torch.Tensor:
         """
         Args:
@@ -38,10 +41,12 @@ class Joiner(nn.Module):
             Output from the encoder. Its shape is (N, T, self.input_dim).
           decoder_out:
             Output from the decoder. Its shape is (N, U, self.input_dim).
-          unused:
+          unused_encoder_out_len:
             This is a placeholder so that we can reuse
             transducer_stateless/beam_search.py in this folder as that
             script assumes the joiner networks accepts 4 inputs.
+          unused_decoder_out_len:
+            Just a placeholder.
         Returns:
           Return a tensor of shape (N, T, U, self.output_dim).
         """

egs/librispeech/ASR/transducer_stateless2/test_model.py

@@ -0,0 +1,49 @@
+#!/usr/bin/env python3
+# Copyright    2022  Xiaomi Corp.        (authors: Fangjun Kuang)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+"""
+To run this file, do:
+
+    cd icefall/egs/librispeech/ASR
+    python ./transducer_stateless2/test_model.py
+"""
+
+import torch
+from train import get_params, get_transducer_model
+
+
+def test_model():
+    params = get_params()
+    params.vocab_size = 500
+    params.blank_id = 0
+    params.context_size = 2
+
+    model = get_transducer_model(params)
+
+    num_param = sum([p.numel() for p in model.parameters()])
+    print(f"Number of model parameters: {num_param}")
+    model.__class__.forward = torch.jit.ignore(model.__class__.forward)
+    torch.jit.script(model)
+
+
+def main():
+    test_model()
+
+
+if __name__ == "__main__":
+    main()

egs/librispeech/ASR/transducer_stateless2/train.py

@@ -511,7 +511,7 @@ def train_one_epoch(
         loss.backward()
         clip_grad_norm_(model.parameters(), 5.0, 2.0)
         optimizer.step()
-        if params.print_diagnostics and batch_idx == 5:
+        if params.print_diagnostics and batch_idx == 30:
             return
 
         if batch_idx % params.log_interval == 0:
@@ -623,10 +623,7 @@ def run(rank, world_size, args):
     librispeech = LibriSpeechAsrDataModule(args)
 
     if params.print_diagnostics:
-        opts = diagnostics.TensorDiagnosticOptions(
+        diagnostic = diagnostics.attach_diagnostics(model)
-            2 ** 22
-        )  # allow 4 megabytes per sub-module
-        diagnostic = diagnostics.attach_diagnostics(model, opts)
 
     train_cuts = librispeech.train_clean_100_cuts()
     if params.full_libri:

egs/librispeech/ASR/transducer_stateless_multi_datasets/export.py

@@ -184,8 +184,6 @@ def main():
     args = get_parser().parse_args()
     args.exp_dir = Path(args.exp_dir)
 
-    assert args.jit is False, "Support torchscript will be added later"
-
     params = get_params()
     params.update(vars(args))
 
@@ -229,6 +227,11 @@ def main():
     model.eval()
 
     if params.jit:
+        # We won't use the forward() method of the model in C++, so just ignore
+        # it here.
+        # Otherwise, one of its arguments is a ragged tensor and is not
+        # torch scriptabe.
+        model.__class__.forward = torch.jit.ignore(model.__class__.forward)
         logging.info("Using torch.jit.script")
         model = torch.jit.script(model)
         filename = params.exp_dir / "cpu_jit.pt"

egs/spgispeech/ASR/README.md

@@ -0,0 +1,32 @@
+# SPGISpeech
+
+SPGISpeech consists of 5,000 hours of recorded company earnings calls and their respective
+transcriptions. The original calls were split into slices ranging from 5 to 15 seconds in
+length to allow easy training for speech recognition systems. Calls represent a broad
+cross-section of international business English; SPGISpeech contains approximately 50,000
+speakers, one of the largest numbers of any speech corpus, and offers a variety of L1 and
+L2 English accents. The format of each WAV file is single channel, 16kHz, 16 bit audio.
+
+Transcription text represents the output of several stages of manual post-processing.
+As such, the text contains polished English orthography following a detailed style guide,
+including proper casing, punctuation, and denormalized non-standard words such as numbers
+and acronyms, making SPGISpeech suited for training fully formatted end-to-end models.
+
+Official reference:
+
+O’Neill, P.K., Lavrukhin, V., Majumdar, S., Noroozi, V., Zhang, Y., Kuchaiev, O., Balam,
+J., Dovzhenko, Y., Freyberg, K., Shulman, M.D., Ginsburg, B., Watanabe, S., & Kucsko, G.
+(2021). SPGISpeech: 5, 000 hours of transcribed financial audio for fully formatted
+end-to-end speech recognition. ArXiv, abs/2104.02014.
+
+ArXiv link: https://arxiv.org/abs/2104.02014
+
+## Performance Record
+
+| Decoding method           | val WER    | val CER |
+|---------------------------|------------|---------|
+| greedy search             | 2.40       |  0.99   |
+| modified beam search      | 2.24       |  0.91   |
+| fast beam search          | 2.35       |  0.97   |
+
+See [RESULTS](/egs/spgispeech/ASR/RESULTS.md) for details.

egs/spgispeech/ASR/RESULTS.md

@@ -0,0 +1,73 @@
+## Results
+
+### SPGISpeech BPE training results (Pruned Transducer)
+
+#### 2022-05-11
+
+#### Conformer encoder + embedding decoder
+
+Conformer encoder + non-current decoder. The decoder
+contains only an embedding layer, a Conv1d (with kernel size 2) and a linear
+layer (to transform tensor dim).
+
+The WERs are
+
+|                           | dev | val | comment                                  |
+|---------------------------|------------|------------|------------------------------------------|
+| greedy search             | 2.46       | 2.40       | --avg-last-n 10 --max-duration 500 |
+| modified beam search      | 2.28       | 2.24       | --avg-last-n 10 --max-duration 500 --beam-size 4 |
+| fast beam search          | 2.38       | 2.35       | --avg-last-n 10 --max-duration 500 --beam-size 4 --max-contexts 4 --max-states 8 |
+
+**NOTE:** SPGISpeech transcripts can be prepared in `ortho` or `norm` ways, which refer to whether the
+transcripts are orthographic or normalized. These WERs correspond to the normalized transcription
+scenario.
+
+The training command for reproducing is given below:
+
+```
+export CUDA_VISIBLE_DEVICES="0,1,2,3,4,5,6,7"
+
+./pruned_transducer_stateless2/train.py \
+  --world-size 8 \
+  --num-epochs 20 \
+  --start-epoch 0 \
+  --exp-dir pruned_transducer_stateless2/exp \
+  --max-duration 200 \
+  --prune-range 5 \
+  --lr-factor 5 \
+  --lm-scale 0.25 \
+  --use-fp16 True
+```
+
+The decoding command is:
+```
+# greedy search
+./pruned_transducer_stateless2/decode.py \
+        --iter 696000 --avg 10 \
+        --exp-dir ./pruned_transducer_stateless2/exp \
+        --max-duration 100 \
+        --decoding-method greedy_search
+
+# modified beam search
+./pruned_transducer_stateless2/decode.py \
+        --iter 696000 --avg 10 \
+        --exp-dir ./pruned_transducer_stateless2/exp \
+        --max-duration 100 \
+        --decoding-method modified_beam_search \
+        --beam-size 4
+
+# fast beam search
+./pruned_transducer_stateless2/decode.py \
+        --iter 696000 --avg 10 \
+        --exp-dir ./pruned_transducer_stateless2/exp \
+        --max-duration 1500 \
+        --decoding-method fast_beam_search \
+        --beam 4 \
+        --max-contexts 4 \
+        --max-states 8
+```
+
+Pretrained model is available at <https://huggingface.co/desh2608/icefall-asr-spgispeech-pruned-transducer-stateless2>
+
+The tensorboard training log can be found at
+<https://tensorboard.dev/experiment/ExSoBmrPRx6liMTGLu0Tgw/#scalars>

egs/spgispeech/ASR/local/compile_hlg.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/local/compile_hlg.py

egs/spgispeech/ASR/local/compute_fbank_musan.py

@@ -0,0 +1,107 @@
+#!/usr/bin/env python3
+# Copyright    2021  Xiaomi Corp.        (authors: Fangjun Kuang)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+"""
+This file computes fbank features of the musan dataset.
+It looks for manifests in the directory data/manifests.
+
+The generated fbank features are saved in data/fbank.
+"""
+
+import logging
+from pathlib import Path
+
+import torch
+from lhotse import CutSet, LilcomChunkyWriter, combine
+from lhotse.features.kaldifeat import (
+    KaldifeatFbank,
+    KaldifeatFbankConfig,
+    KaldifeatFrameOptions,
+    KaldifeatMelOptions,
+)
+from lhotse.recipes.utils import read_manifests_if_cached
+
+# Torch's multithreaded behavior needs to be disabled or
+# it wastes a lot of CPU and slow things down.
+# Do this outside of main() in case it needs to take effect
+# even when we are not invoking the main (e.g. when spawning subprocesses).
+torch.set_num_threads(1)
+torch.set_num_interop_threads(1)
+
+
+def compute_fbank_musan():
+    src_dir = Path("data/manifests")
+    output_dir = Path("data/fbank")
+
+    sampling_rate = 16000
+    num_mel_bins = 80
+
+    extractor = KaldifeatFbank(
+        KaldifeatFbankConfig(
+            frame_opts=KaldifeatFrameOptions(sampling_rate=sampling_rate),
+            mel_opts=KaldifeatMelOptions(num_bins=num_mel_bins),
+            device="cuda",
+        )
+    )
+
+    dataset_parts = (
+        "music",
+        "speech",
+        "noise",
+    )
+    manifests = read_manifests_if_cached(
+        dataset_parts=dataset_parts, output_dir=src_dir
+    )
+    assert manifests is not None
+
+    musan_cuts_path = src_dir / "cuts_musan.jsonl.gz"
+
+    if musan_cuts_path.is_file():
+        logging.info(f"{musan_cuts_path} already exists - skipping")
+        return
+
+    logging.info("Extracting features for Musan")
+
+    # create chunks of Musan with duration 5 - 10 seconds
+    musan_cuts = (
+        CutSet.from_manifests(
+            recordings=combine(
+                part["recordings"] for part in manifests.values()
+            )
+        )
+        .cut_into_windows(10.0)
+        .filter(lambda c: c.duration > 5)
+        .compute_and_store_features_batch(
+            extractor=extractor,
+            storage_path=output_dir / "feats_musan",
+            batch_duration=500,
+            num_workers=4,
+            storage_type=LilcomChunkyWriter,
+        )
+    )
+
+    logging.info(f"Saving to {musan_cuts_path}")
+    musan_cuts.to_file(musan_cuts_path)
+
+
+if __name__ == "__main__":
+    formatter = (
+        "%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s"
+    )
+    logging.basicConfig(format=formatter, level=logging.INFO)
+    compute_fbank_musan()

egs/spgispeech/ASR/local/compute_fbank_spgispeech.py

@@ -0,0 +1,151 @@
+#!/usr/bin/env python3
+# Copyright    2022  Johns Hopkins University        (authors: Desh Raj)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+"""
+This file computes fbank features of the SPGISpeech dataset.
+It looks for manifests in the directory data/manifests.
+
+The generated fbank features are saved in data/fbank.
+"""
+import argparse
+import logging
+from pathlib import Path
+
+import torch
+from lhotse import LilcomChunkyWriter, load_manifest_lazy
+from lhotse.features.kaldifeat import (
+    KaldifeatFbank,
+    KaldifeatFbankConfig,
+    KaldifeatFrameOptions,
+    KaldifeatMelOptions,
+)
+
+# Torch's multithreaded behavior needs to be disabled or
+# it wastes a lot of CPU and slow things down.
+# Do this outside of main() in case it needs to take effect
+# even when we are not invoking the main (e.g. when spawning subprocesses).
+torch.set_num_threads(1)
+torch.set_num_interop_threads(1)
+
+
+def get_args():
+    parser = argparse.ArgumentParser()
+    parser.add_argument(
+        "--num-splits",
+        type=int,
+        default=20,
+        help="Number of splits for the train set.",
+    )
+    parser.add_argument(
+        "--start",
+        type=int,
+        default=0,
+        help="Start index of the train set split.",
+    )
+    parser.add_argument(
+        "--stop",
+        type=int,
+        default=-1,
+        help="Stop index of the train set split.",
+    )
+    parser.add_argument(
+        "--test",
+        action="store_true",
+        help="If set, only compute features for the dev and val set.",
+    )
+    parser.add_argument(
+        "--train",
+        action="store_true",
+        help="If set, only compute features for the train set.",
+    )
+
+    return parser.parse_args()
+
+
+def compute_fbank_spgispeech(args):
+    assert args.train or args.test, "Either train or test must be set."
+
+    src_dir = Path("data/manifests")
+    output_dir = Path("data/fbank")
+
+    sampling_rate = 16000
+    num_mel_bins = 80
+
+    extractor = KaldifeatFbank(
+        KaldifeatFbankConfig(
+            frame_opts=KaldifeatFrameOptions(sampling_rate=sampling_rate),
+            mel_opts=KaldifeatMelOptions(num_bins=num_mel_bins),
+            device="cuda",
+        )
+    )
+
+    if args.train:
+        logging.info("Processing train")
+        cut_set = load_manifest_lazy(src_dir / "cuts_train_raw.jsonl.gz")
+        chunk_size = len(cut_set) // args.num_splits
+        cut_sets = cut_set.split_lazy(
+            output_dir=src_dir / f"cuts_train_raw_split{args.num_splits}",
+            chunk_size=chunk_size,
+        )
+        start = args.start
+        stop = (
+            min(args.stop, args.num_splits)
+            if args.stop > 0
+            else args.num_splits
+        )
+        num_digits = len(str(args.num_splits))
+        for i in range(start, stop):
+            idx = f"{i + 1}".zfill(num_digits)
+            cuts_train_idx_path = src_dir / f"cuts_train_{idx}.jsonl.gz"
+            logging.info(f"Processing train split {i}")
+            cs = cut_sets[i].compute_and_store_features_batch(
+                extractor=extractor,
+                storage_path=output_dir / f"feats_train_{idx}",
+                batch_duration=500,
+                num_workers=4,
+                storage_type=LilcomChunkyWriter,
+            )
+            cs.to_file(cuts_train_idx_path)
+
+    if args.test:
+        for partition in ["dev", "val"]:
+            if (output_dir / f"cuts_{partition}.jsonl.gz").is_file():
+                logging.info(f"{partition} already exists - skipping.")
+                continue
+            logging.info(f"Processing {partition}")
+            cut_set = load_manifest_lazy(
+                src_dir / f"cuts_{partition}_raw.jsonl.gz"
+            )
+            cut_set = cut_set.compute_and_store_features_batch(
+                extractor=extractor,
+                storage_path=output_dir / f"feats_{partition}",
+                manifest_path=src_dir / f"cuts_{partition}.jsonl.gz",
+                batch_duration=500,
+                num_workers=4,
+                storage_type=LilcomChunkyWriter,
+            )
+
+
+if __name__ == "__main__":
+    formatter = (
+        "%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s"
+    )
+    logging.basicConfig(format=formatter, level=logging.INFO)
+
+    args = get_args()
+    compute_fbank_spgispeech(args)

egs/spgispeech/ASR/local/prepare_lang.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/local/prepare_lang.py

egs/spgispeech/ASR/local/prepare_lang_bpe.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/local/prepare_lang_bpe.py

egs/spgispeech/ASR/local/prepare_splits.py

@@ -0,0 +1,81 @@
+#!/usr/bin/env python3
+# Copyright    2022  Johns Hopkins University        (authors: Desh Raj)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+"""
+This file splits the training set into train and dev sets.
+"""
+import logging
+from pathlib import Path
+
+import torch
+from lhotse import CutSet
+from lhotse.recipes.utils import read_manifests_if_cached
+
+# Torch's multithreaded behavior needs to be disabled or
+# it wastes a lot of CPU and slow things down.
+# Do this outside of main() in case it needs to take effect
+# even when we are not invoking the main (e.g. when spawning subprocesses).
+torch.set_num_threads(1)
+torch.set_num_interop_threads(1)
+
+
+def split_spgispeech_train():
+    src_dir = Path("data/manifests")
+
+    manifests = read_manifests_if_cached(
+        dataset_parts=["train", "val"],
+        output_dir=src_dir,
+        prefix="spgispeech",
+        suffix="jsonl.gz",
+        lazy=True,
+    )
+    assert manifests is not None
+
+    train_dev_cuts = CutSet.from_manifests(
+        recordings=manifests["train"]["recordings"],
+        supervisions=manifests["train"]["supervisions"],
+    )
+    dev_cuts = train_dev_cuts.subset(first=4000)
+    train_cuts = train_dev_cuts.filter(lambda c: c not in dev_cuts)
+
+    # Add speed perturbation
+    train_cuts = (
+        train_cuts
+        + train_cuts.perturb_speed(0.9)
+        + train_cuts.perturb_speed(1.1)
+    )
+
+    # Write the manifests to disk.
+    train_cuts.to_file(src_dir / "cuts_train_raw.jsonl.gz")
+    dev_cuts.to_file(src_dir / "cuts_dev_raw.jsonl.gz")
+
+    # Also write the val set to disk.
+    val_cuts = CutSet.from_manifests(
+        recordings=manifests["val"]["recordings"],
+        supervisions=manifests["val"]["supervisions"],
+    )
+    val_cuts.to_file(src_dir / "cuts_val_raw.jsonl.gz")
+
+
+if __name__ == "__main__":
+    formatter = (
+        "%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s"
+    )
+    logging.basicConfig(format=formatter, level=logging.INFO)
+
+    split_spgispeech_train()

egs/spgispeech/ASR/local/train_bpe_model.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/local/train_bpe_model.py

egs/spgispeech/ASR/prepare.sh

@@ -0,0 +1,196 @@
+#!/usr/bin/env bash
+
+set -eou pipefail
+
+nj=20
+stage=-1
+stop_stage=100
+
+# We assume dl_dir (download dir) contains the following
+# directories and files. If not, they will be downloaded
+# by this script automatically.
+#
+#  - $dl_dir/spgispeech
+#      You can find train.csv, val.csv, train, and val in this directory, which belong
+#      to the SPGISpeech dataset.
+#
+#  - $dl_dir/musan
+#      This directory contains the following directories downloaded from
+#       http://www.openslr.org/17/
+#
+#     - music
+#     - noise
+#     - speech
+dl_dir=$PWD/download
+
+. shared/parse_options.sh || exit 1
+
+# vocab size for sentence piece models.
+# It will generate data/lang_bpe_xxx,
+# data/lang_bpe_yyy if the array contains xxx, yyy
+vocab_sizes=(
+  500
+)
+
+# All files generated by this script are saved in "data".
+# You can safely remove "data" and rerun this script to regenerate it.
+mkdir -p data
+
+log() {
+  # This function is from espnet
+  local fname=${BASH_SOURCE[1]##*/}
+  echo -e "$(date '+%Y-%m-%d %H:%M:%S') (${fname}:${BASH_LINENO[0]}:${FUNCNAME[1]}) $*"
+}
+
+log "dl_dir: $dl_dir"
+
+if [ $stage -le 0 ] && [ $stop_stage -ge 0 ]; then
+  log "Stage 0: Download data"
+
+  # If you have pre-downloaded it to /path/to/spgispeech,
+  # you can create a symlink
+  #
+  #   ln -sfv /path/to/spgispeech $dl_dir/spgispeech
+  #
+  if [ ! -d $dl_dir/spgispeech/train.csv ]; then
+    lhotse download spgispeech $dl_dir
+  fi
+
+  # If you have pre-downloaded it to /path/to/musan,
+  # you can create a symlink
+  #
+  #   ln -sfv /path/to/musan $dl_dir/
+  #
+  if [ ! -d $dl_dir/musan ]; then
+    lhotse download musan $dl_dir
+  fi
+fi
+
+if [ $stage -le 1 ] && [ $stop_stage -ge 1 ]; then
+  log "Stage 1: Prepare SPGISpeech manifest (may take ~1h)"
+  # We assume that you have downloaded the SPGISpeech corpus
+  # to $dl_dir/spgispeech. We perform text normalization for the transcripts.
+  mkdir -p data/manifests
+  lhotse prepare spgispeech -j $nj --normalize-text $dl_dir/spgispeech data/manifests
+fi
+
+if [ $stage -le 2 ] && [ $stop_stage -ge 2 ]; then
+  log "Stage 2: Prepare musan manifest"
+  # We assume that you have downloaded the musan corpus
+  # to data/musan
+  mkdir -p data/manifests
+  lhotse prepare musan $dl_dir/musan data/manifests
+  lhotse combine data/manifests/recordings_{music,speech,noise}.json data/manifests/recordings_musan.jsonl.gz
+  lhotse cut simple -r data/manifests/recordings_musan.jsonl.gz data/manifests/cuts_musan_raw.jsonl.gz
+fi
+
+if [ $stage -le 3 ] && [ $stop_stage -ge 3 ]; then
+  log "Stage 3: Split train into train and dev and create cut sets."
+  python local/prepare_splits.py
+fi
+
+if [ $stage -le 4 ] && [ $stop_stage -ge 4 ]; then
+  log "Stage 4: Compute fbank features for spgispeech dev and val"
+  mkdir -p data/fbank
+  python local/compute_fbank_spgispeech.py --test
+fi
+
+if [ $stage -le 5 ] && [ $stop_stage -ge 5 ]; then
+  log "Stage 5: Compute fbank features for train"
+  mkdir -p data/fbank
+  python local/compute_fbank_spgispeech.py --train --num-splits 20
+
+  log "Combine features from train splits (may take ~1h)"
+  if [ ! -f data/manifests/cuts_train.jsonl.gz ]; then
+    pieces=$(find data/manifests -name "cuts_train_[0-9]*.jsonl.gz")
+    lhotse combine $pieces data/manifests/cuts_train.jsonl.gz
+  fi
+  gunzip -c data/manifests/train_cuts.jsonl.gz | shuf | gzip -c > data/manifests/train_cuts_shuf.jsonl.gz
+fi
+
+if [ $stage -le 6 ] && [ $stop_stage -ge 6 ]; then
+  log "Stage 6: Compute fbank features for musan"
+  mkdir -p data/fbank
+  python local/compute_fbank_musan.py
+fi
+
+if [ $stage -le 7 ] && [ $stop_stage -ge 7 ]; then
+  log "Stage 7: Dump transcripts for LM training"
+  mkdir -p data/lm
+  gunzip -c data/manifests/cuts_train_raw.jsonl.gz \
+    | jq '.supervisions[0].text' \
+    | sed 's:"::g' \
+    > data/lm/transcript_words.txt
+fi
+
+if [ $stage -le 8 ] && [ $stop_stage -ge 8 ]; then
+  log "Stage 8: Prepare BPE based lang"
+
+  for vocab_size in ${vocab_sizes[@]}; do
+    lang_dir=data/lang_bpe_${vocab_size}
+    mkdir -p $lang_dir
+
+    # Add special words to words.txt
+    echo "<eps> 0" > $lang_dir/words.txt
+    echo "!SIL 1" >> $lang_dir/words.txt
+    echo "[UNK] 2" >> $lang_dir/words.txt
+
+    # Add regular words to words.txt
+    gunzip -c data/manifests/cuts_train_raw.jsonl.gz \
+      | jq '.supervisions[0].text' \
+      | sed 's:"::g' \
+      | sed 's: :\n:g' \
+      | sort \
+      | uniq \
+      | sed '/^$/d' \
+      | awk '{print $0,NR+2}' \
+      >> $lang_dir/words.txt
+
+    # Add remaining special word symbols expected by LM scripts.
+    num_words=$(cat $lang_dir/words.txt | wc -l)
+    echo "<s> ${num_words}" >> $lang_dir/words.txt
+    num_words=$(cat $lang_dir/words.txt | wc -l)
+    echo "</s> ${num_words}" >> $lang_dir/words.txt
+    num_words=$(cat $lang_dir/words.txt | wc -l)
+    echo "#0 ${num_words}" >> $lang_dir/words.txt
+
+    ./local/train_bpe_model.py \
+      --lang-dir $lang_dir \
+      --vocab-size $vocab_size \
+      --transcript data/lm/transcript_words.txt
+
+    if [ ! -f $lang_dir/L_disambig.pt ]; then
+      ./local/prepare_lang_bpe.py --lang-dir $lang_dir
+    fi
+  done
+fi
+
+if [ $stage -le 9 ] && [ $stop_stage -ge 9 ]; then
+  log "Stage 9: Train LM"
+  lm_dir=data/lm
+
+  if [ ! -f $lm_dir/G.arpa ]; then
+    ./shared/make_kn_lm.py \
+      -ngram-order 3 \
+      -text $lm_dir/transcript_words.txt \
+      -lm $lm_dir/G.arpa
+  fi
+
+  if [ ! -f $lm_dir/G_3_gram.fst.txt ]; then
+    python3 -m kaldilm \
+      --read-symbol-table="data/lang_phone/words.txt" \
+      --disambig-symbol='#0' \
+      --max-order=3 \
+      $lm_dir/G.arpa > $lm_dir/G_3_gram.fst.txt
+  fi
+fi
+
+if [ $stage -le 10 ] && [ $stop_stage -ge 10 ]; then
+  log "Stage 10: Compile HLG"
+  ./local/compile_hlg.py --lang-dir data/lang_phone
+
+  for vocab_size in ${vocab_sizes[@]}; do
+    lang_dir=data/lang_bpe_${vocab_size}
+    ./local/compile_hlg.py --lang-dir $lang_dir
+  done
+fi

egs/spgispeech/ASR/pruned_transducer_stateless2/asr_datamodule.py

@@ -0,0 +1,366 @@
+# Copyright      2021  Piotr Żelasko
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+
+import argparse
+import logging
+from functools import lru_cache
+from pathlib import Path
+from typing import Any, Dict, Optional
+
+import torch
+from lhotse import CutSet, Fbank, FbankConfig, load_manifest, load_manifest_lazy
+from lhotse.dataset import (
+    CutConcatenate,
+    CutMix,
+    DynamicBucketingSampler,
+    K2SpeechRecognitionDataset,
+    PrecomputedFeatures,
+    SpecAugment,
+)
+from lhotse.dataset.input_strategies import OnTheFlyFeatures
+from lhotse.utils import fix_random_seed
+from torch.utils.data import DataLoader
+from tqdm import tqdm
+
+from icefall.utils import str2bool
+
+
+class _SeedWorkers:
+    def __init__(self, seed: int):
+        self.seed = seed
+
+    def __call__(self, worker_id: int):
+        fix_random_seed(self.seed + worker_id)
+
+
+class SPGISpeechAsrDataModule:
+    """
+    DataModule for k2 ASR experiments.
+    It assumes there is always one train and valid dataloader,
+    but there can be multiple test dataloaders (e.g. LibriSpeech test-clean
+    and test-other).
+    It contains all the common data pipeline modules used in ASR
+    experiments, e.g.:
+    - dynamic batch size,
+    - bucketing samplers,
+    - cut concatenation,
+    - augmentation,
+    - on-the-fly feature extraction
+    This class should be derived for specific corpora used in ASR tasks.
+    """
+
+    def __init__(self, args: argparse.Namespace):
+        self.args = args
+
+    @classmethod
+    def add_arguments(cls, parser: argparse.ArgumentParser):
+        group = parser.add_argument_group(
+            title="ASR data related options",
+            description="These options are used for the preparation of "
+            "PyTorch DataLoaders from Lhotse CutSet's -- they control the "
+            "effective batch sizes, sampling strategies, applied data "
+            "augmentations, etc.",
+        )
+        group.add_argument(
+            "--manifest-dir",
+            type=Path,
+            default=Path("data/manifests"),
+            help="Path to directory with train/valid/test cuts.",
+        )
+        group.add_argument(
+            "--enable-musan",
+            type=str2bool,
+            default=True,
+            help="When enabled, select noise from MUSAN and mix it "
+            "with training dataset. ",
+        )
+        group.add_argument(
+            "--concatenate-cuts",
+            type=str2bool,
+            default=False,
+            help="When enabled, utterances (cuts) will be concatenated "
+            "to minimize the amount of padding.",
+        )
+        group.add_argument(
+            "--duration-factor",
+            type=float,
+            default=1.0,
+            help="Determines the maximum duration of a concatenated cut "
+            "relative to the duration of the longest cut in a batch.",
+        )
+        group.add_argument(
+            "--gap",
+            type=float,
+            default=1.0,
+            help="The amount of padding (in seconds) inserted between "
+            "concatenated cuts. This padding is filled with noise when "
+            "noise augmentation is used.",
+        )
+        group.add_argument(
+            "--max-duration",
+            type=int,
+            default=100.0,
+            help="Maximum pooled recordings duration (seconds) in a "
+            "single batch. You can reduce it if it causes CUDA OOM.",
+        )
+        group.add_argument(
+            "--num-buckets",
+            type=int,
+            default=30,
+            help="The number of buckets for the BucketingSampler"
+            "(you might want to increase it for larger datasets).",
+        )
+        group.add_argument(
+            "--on-the-fly-feats",
+            type=str2bool,
+            default=False,
+            help="When enabled, use on-the-fly cut mixing and feature "
+            "extraction. Will drop existing precomputed feature manifests "
+            "if available.",
+        )
+        group.add_argument(
+            "--shuffle",
+            type=str2bool,
+            default=True,
+            help="When enabled (=default), the examples will be "
+            "shuffled for each epoch.",
+        )
+
+        group.add_argument(
+            "--num-workers",
+            type=int,
+            default=8,
+            help="The number of training dataloader workers that "
+            "collect the batches.",
+        )
+        group.add_argument(
+            "--enable-spec-aug",
+            type=str2bool,
+            default=True,
+            help="When enabled, use SpecAugment for training dataset.",
+        )
+        group.add_argument(
+            "--spec-aug-time-warp-factor",
+            type=int,
+            default=80,
+            help="Used only when --enable-spec-aug is True. "
+            "It specifies the factor for time warping in SpecAugment. "
+            "Larger values mean more warping. "
+            "A value less than 1 means to disable time warp.",
+        )
+
+    def train_dataloaders(
+        self,
+        cuts_train: CutSet,
+        sampler_state_dict: Optional[Dict[str, Any]] = None,
+    ) -> DataLoader:
+        """
+        Args:
+          cuts_train:
+            CutSet for training.
+          sampler_state_dict:
+            The state dict for the training sampler.
+        """
+        logging.info("About to get Musan cuts")
+        cuts_musan = load_manifest(
+            self.args.manifest_dir / "cuts_musan.jsonl.gz"
+        )
+
+        transforms = []
+        if self.args.enable_musan:
+            logging.info("Enable MUSAN")
+            transforms.append(
+                CutMix(
+                    cuts=cuts_musan, prob=0.5, snr=(10, 20), preserve_id=True
+                )
+            )
+        else:
+            logging.info("Disable MUSAN")
+
+        if self.args.concatenate_cuts:
+            logging.info(
+                f"Using cut concatenation with duration factor "
+                f"{self.args.duration_factor} and gap {self.args.gap}."
+            )
+            # Cut concatenation should be the first transform in the list,
+            # so that if we e.g. mix noise in, it will fill the gaps between
+            # different utterances.
+            transforms = [
+                CutConcatenate(
+                    duration_factor=self.args.duration_factor, gap=self.args.gap
+                )
+            ] + transforms
+
+        input_transforms = []
+        if self.args.enable_spec_aug:
+            logging.info("Enable SpecAugment")
+            logging.info(
+                f"Time warp factor: {self.args.spec_aug_time_warp_factor}"
+            )
+            input_transforms.append(
+                SpecAugment(
+                    time_warp_factor=self.args.spec_aug_time_warp_factor,
+                    num_frame_masks=2,
+                    features_mask_size=27,
+                    num_feature_masks=2,
+                    frames_mask_size=100,
+                )
+            )
+        else:
+            logging.info("Disable SpecAugment")
+
+        logging.info("About to create train dataset")
+        if self.args.on_the_fly_feats:
+            train = K2SpeechRecognitionDataset(
+                cut_transforms=transforms,
+                input_strategy=OnTheFlyFeatures(
+                    Fbank(FbankConfig(num_mel_bins=80))
+                ),
+                input_transforms=input_transforms,
+            )
+        else:
+            train = K2SpeechRecognitionDataset(
+                cut_transforms=transforms,
+                input_transforms=input_transforms,
+            )
+
+        logging.info("Using DynamicBucketingSampler.")
+        train_sampler = DynamicBucketingSampler(
+            cuts_train,
+            max_duration=self.args.max_duration,
+            shuffle=False,
+            num_buckets=self.args.num_buckets,
+            drop_last=True,
+        )
+        logging.info("About to create train dataloader")
+
+        if sampler_state_dict is not None:
+            logging.info("Loading sampler state dict")
+            train_sampler.load_state_dict(sampler_state_dict)
+
+        # 'seed' is derived from the current random state, which will have
+        # previously been set in the main process.
+        seed = torch.randint(0, 100000, ()).item()
+        worker_init_fn = _SeedWorkers(seed)
+
+        train_dl = DataLoader(
+            train,
+            sampler=train_sampler,
+            batch_size=None,
+            num_workers=self.args.num_workers,
+            persistent_workers=False,
+            worker_init_fn=worker_init_fn,
+        )
+
+        return train_dl
+
+    def valid_dataloaders(self, cuts_valid: CutSet) -> DataLoader:
+
+        transforms = []
+        if self.args.concatenate_cuts:
+            transforms = [
+                CutConcatenate(
+                    duration_factor=self.args.duration_factor, gap=self.args.gap
+                )
+            ] + transforms
+
+        logging.info("About to create dev dataset")
+        if self.args.on_the_fly_feats:
+            validate = K2SpeechRecognitionDataset(
+                cut_transforms=transforms,
+                input_strategy=OnTheFlyFeatures(
+                    Fbank(FbankConfig(num_mel_bins=80))
+                ),
+            )
+        else:
+            validate = K2SpeechRecognitionDataset(
+                cut_transforms=transforms,
+            )
+        valid_sampler = DynamicBucketingSampler(
+            cuts_valid,
+            max_duration=self.args.max_duration,
+            shuffle=False,
+        )
+        logging.info("About to create dev dataloader")
+        valid_dl = DataLoader(
+            validate,
+            sampler=valid_sampler,
+            batch_size=None,
+            num_workers=2,
+            persistent_workers=False,
+        )
+
+        return valid_dl
+
+    def test_dataloaders(self, cuts: CutSet) -> DataLoader:
+        logging.debug("About to create test dataset")
+        test = K2SpeechRecognitionDataset(
+            input_strategy=OnTheFlyFeatures(Fbank(FbankConfig(num_mel_bins=80)))
+            if self.args.on_the_fly_feats
+            else PrecomputedFeatures(),
+        )
+        sampler = DynamicBucketingSampler(
+            cuts, max_duration=self.args.max_duration, shuffle=False
+        )
+        logging.debug("About to create test dataloader")
+        test_dl = DataLoader(
+            test,
+            batch_size=None,
+            sampler=sampler,
+            num_workers=self.args.num_workers,
+        )
+        return test_dl
+
+    @lru_cache()
+    def train_cuts(self) -> CutSet:
+        logging.info("About to get SPGISpeech train cuts")
+        return load_manifest_lazy(
+            self.args.manifest_dir / "cuts_train_shuf.jsonl.gz"
+        )
+
+    @lru_cache()
+    def dev_cuts(self) -> CutSet:
+        logging.info("About to get SPGISpeech dev cuts")
+        return load_manifest_lazy(self.args.manifest_dir / "cuts_dev.jsonl.gz")
+
+    @lru_cache()
+    def val_cuts(self) -> CutSet:
+        logging.info("About to get SPGISpeech val cuts")
+        return load_manifest_lazy(self.args.manifest_dir / "cuts_val.jsonl.gz")
+
+
+def test():
+    parser = argparse.ArgumentParser()
+    SPGISpeechAsrDataModule.add_arguments(parser)
+    args = parser.parse_args()
+    adm = SPGISpeechAsrDataModule(args)
+
+    cuts = adm.train_cuts()
+    dl = adm.train_dataloaders(cuts)
+    for i, batch in tqdm(enumerate(dl)):
+        if i == 100:
+            break
+
+    cuts = adm.dev_cuts()
+    dl = adm.valid_dataloaders(cuts)
+    for i, batch in tqdm(enumerate(dl)):
+        if i == 100:
+            break
+
+
+if __name__ == "__main__":
+    test()

egs/spgispeech/ASR/pruned_transducer_stateless2/beam_search.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/beam_search.py

egs/spgispeech/ASR/pruned_transducer_stateless2/conformer.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/conformer.py

egs/spgispeech/ASR/pruned_transducer_stateless2/decode.py

@@ -0,0 +1,594 @@
+#!/usr/bin/env python3
+#
+# Copyright 2021 Xiaomi Corporation (Author: Fangjun Kuang)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+"""
+Usage:
+(1) greedy search
+./pruned_transducer_stateless2/decode.py \
+        --iter 696000 \
+        --avg 10 \
+        --exp-dir ./pruned_transducer_stateless2/exp \
+        --max-duration 100 \
+        --decoding-method greedy_search
+
+(2) beam search
+./pruned_transducer_stateless2/decode.py \
+        --iter 696000 \
+        --avg 10 \
+        --exp-dir ./pruned_transducer_stateless2/exp \
+        --max-duration 100 \
+        --decoding-method beam_search \
+        --beam-size 4
+
+(3) modified beam search
+./pruned_transducer_stateless2/decode.py \
+        --iter 696000 \
+        --avg 10 \
+        --exp-dir ./pruned_transducer_stateless2/exp \
+        --max-duration 100 \
+        --decoding-method modified_beam_search \
+        --beam-size 4
+
+(4) fast beam search
+./pruned_transducer_stateless2/decode.py \
+        --iter 696000 \
+        --avg 10 \
+        --exp-dir ./pruned_transducer_stateless2/exp \
+        --max-duration 1500 \
+        --decoding-method fast_beam_search \
+        --beam 4 \
+        --max-contexts 4 \
+        --max-states 8
+"""
+
+
+import argparse
+import logging
+from collections import defaultdict
+from pathlib import Path
+from typing import Dict, List, Optional, Tuple
+
+import k2
+import sentencepiece as spm
+import torch
+import torch.nn as nn
+from asr_datamodule import SPGISpeechAsrDataModule
+from beam_search import (
+    beam_search,
+    fast_beam_search_one_best,
+    greedy_search,
+    greedy_search_batch,
+    modified_beam_search,
+)
+from train import get_params, get_transducer_model
+
+from icefall.checkpoint import (
+    average_checkpoints,
+    find_checkpoints,
+    load_checkpoint,
+)
+from icefall.utils import (
+    AttributeDict,
+    setup_logger,
+    store_transcripts,
+    write_error_stats,
+)
+
+
+def get_parser():
+    parser = argparse.ArgumentParser(
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter
+    )
+
+    parser.add_argument(
+        "--epoch",
+        type=int,
+        default=20,
+        help="""It specifies the checkpoint to use for decoding.
+        Note: Epoch counts from 0.
+        You can specify --avg to use more checkpoints for model averaging.""",
+    )
+
+    parser.add_argument(
+        "--iter",
+        type=int,
+        default=0,
+        help="""If positive, --epoch is ignored and it
+        will use the checkpoint exp_dir/checkpoint-iter.pt.
+        You can specify --avg to use more checkpoints for model averaging.
+        """,
+    )
+
+    parser.add_argument(
+        "--avg",
+        type=int,
+        default=10,
+        help="Number of checkpoints to average. Automatically select "
+        "consecutive checkpoints before the checkpoint specified by "
+        "'--epoch' and '--iter'",
+    )
+
+    parser.add_argument(
+        "--exp-dir",
+        type=str,
+        default="pruned_transducer_stateless2/exp",
+        help="The experiment dir",
+    )
+
+    parser.add_argument(
+        "--bpe-model",
+        type=str,
+        default="data/lang_bpe_500/bpe.model",
+        help="Path to the BPE model",
+    )
+
+    parser.add_argument(
+        "--decoding-method",
+        type=str,
+        default="greedy_search",
+        help="""Possible values are:
+          - greedy_search
+          - beam_search
+          - modified_beam_search
+          - fast_beam_search
+        """,
+    )
+
+    parser.add_argument(
+        "--beam-size",
+        type=int,
+        default=4,
+        help="""An interger indicating how many candidates we will keep for each
+        frame. Used only when --decoding-method is beam_search or
+        modified_beam_search.""",
+    )
+
+    parser.add_argument(
+        "--beam",
+        type=float,
+        default=4,
+        help="""A floating point value to calculate the cutoff score during beam
+        search (i.e., `cutoff = max-score - beam`), which is the same as the
+        `beam` in Kaldi.
+        Used only when --decoding-method is fast_beam_search""",
+    )
+
+    parser.add_argument(
+        "--max-contexts",
+        type=int,
+        default=4,
+        help="""Used only when --decoding-method is
+        fast_beam_search""",
+    )
+
+    parser.add_argument(
+        "--max-states",
+        type=int,
+        default=8,
+        help="""Used only when --decoding-method is
+        fast_beam_search""",
+    )
+
+    parser.add_argument(
+        "--context-size",
+        type=int,
+        default=2,
+        help="The context size in the decoder. 1 means bigram; "
+        "2 means tri-gram",
+    )
+    parser.add_argument(
+        "--max-sym-per-frame",
+        type=int,
+        default=1,
+        help="""Maximum number of symbols per frame.
+        Used only when --decoding_method is greedy_search""",
+    )
+
+    return parser
+
+
+def decode_one_batch(
+    params: AttributeDict,
+    model: nn.Module,
+    sp: spm.SentencePieceProcessor,
+    batch: dict,
+    decoding_graph: Optional[k2.Fsa] = None,
+) -> Dict[str, List[List[str]]]:
+    """Decode one batch and return the result in a dict. The dict has the
+    following format:
+
+        - key: It indicates the setting used for decoding. For example,
+               if greedy_search is used, it would be "greedy_search"
+               If beam search with a beam size of 7 is used, it would be
+               "beam_7"
+        - value: It contains the decoding result. `len(value)` equals to
+                 batch size. `value[i]` is the decoding result for the i-th
+                 utterance in the given batch.
+    Args:
+      params:
+        It's the return value of :func:`get_params`.
+      model:
+        The neural model.
+      sp:
+        The BPE model.
+      batch:
+        It is the return value from iterating
+        `lhotse.dataset.K2SpeechRecognitionDataset`. See its documentation
+        for the format of the `batch`.
+      decoding_graph:
+        The decoding graph. Can be either a `k2.trivial_graph` or HLG, Used
+        only when --decoding_method is fast_beam_search.
+    Returns:
+      Return the decoding result. See above description for the format of
+      the returned dict.
+    """
+    device = model.device
+    feature = batch["inputs"]
+    assert feature.ndim == 3
+
+    feature = feature.to(device)
+    # at entry, feature is (N, T, C)
+
+    supervisions = batch["supervisions"]
+    feature_lens = supervisions["num_frames"].to(device)
+
+    encoder_out, encoder_out_lens = model.encoder(
+        x=feature, x_lens=feature_lens
+    )
+    hyps = []
+
+    if params.decoding_method == "fast_beam_search":
+        hyp_tokens = fast_beam_search_one_best(
+            model=model,
+            decoding_graph=decoding_graph,
+            encoder_out=encoder_out,
+            encoder_out_lens=encoder_out_lens,
+            beam=params.beam,
+            max_contexts=params.max_contexts,
+            max_states=params.max_states,
+        )
+        for hyp in sp.decode(hyp_tokens):
+            hyps.append(hyp.split())
+    elif (
+        params.decoding_method == "greedy_search"
+        and params.max_sym_per_frame == 1
+    ):
+        hyp_tokens = greedy_search_batch(
+            model=model,
+            encoder_out=encoder_out,
+            encoder_out_lens=encoder_out_lens,
+        )
+        for hyp in sp.decode(hyp_tokens):
+            hyps.append(hyp.split())
+    elif params.decoding_method == "modified_beam_search":
+        hyp_tokens = modified_beam_search(
+            model=model,
+            encoder_out=encoder_out,
+            encoder_out_lens=encoder_out_lens,
+            beam=params.beam_size,
+        )
+        for hyp in sp.decode(hyp_tokens):
+            hyps.append(hyp.split())
+    else:
+        batch_size = encoder_out.size(0)
+
+        for i in range(batch_size):
+            # fmt: off
+            encoder_out_i = encoder_out[i:i+1, :encoder_out_lens[i]]
+            # fmt: on
+            if params.decoding_method == "greedy_search":
+                hyp = greedy_search(
+                    model=model,
+                    encoder_out=encoder_out_i,
+                    max_sym_per_frame=params.max_sym_per_frame,
+                )
+            elif params.decoding_method == "beam_search":
+                hyp = beam_search(
+                    model=model,
+                    encoder_out=encoder_out_i,
+                    beam=params.beam_size,
+                )
+            else:
+                raise ValueError(
+                    f"Unsupported decoding method: {params.decoding_method}"
+                )
+            hyps.append(sp.decode(hyp).split())
+
+    if params.decoding_method == "greedy_search":
+        return {"greedy_search": hyps}
+    elif params.decoding_method == "fast_beam_search":
+        return {
+            (
+                f"beam_{params.beam}_"
+                f"max_contexts_{params.max_contexts}_"
+                f"max_states_{params.max_states}"
+            ): hyps
+        }
+    else:
+        return {f"beam_size_{params.beam_size}": hyps}
+
+
+def decode_dataset(
+    dl: torch.utils.data.DataLoader,
+    params: AttributeDict,
+    model: nn.Module,
+    sp: spm.SentencePieceProcessor,
+    decoding_graph: Optional[k2.Fsa] = None,
+) -> Dict[str, List[Tuple[List[str], List[str]]]]:
+    """Decode dataset.
+
+    Args:
+      dl:
+        PyTorch's dataloader containing the dataset to decode.
+      params:
+        It is returned by :func:`get_params`.
+      model:
+        The neural model.
+      sp:
+        The BPE model.
+      decoding_graph:
+        The decoding graph. Can be either a `k2.trivial_graph` or HLG, Used
+        only when --decoding_method is fast_beam_search.
+    Returns:
+      Return a dict, whose key may be "greedy_search" if greedy search
+      is used, or it may be "beam_7" if beam size of 7 is used.
+      Its value is a list of tuples. Each tuple contains two elements:
+      The first is the reference transcript, and the second is the
+      predicted result.
+    """
+    num_cuts = 0
+
+    try:
+        num_batches = len(dl)
+    except TypeError:
+        num_batches = "?"
+
+    if params.decoding_method == "greedy_search":
+        log_interval = 100
+    else:
+        log_interval = 2
+
+    results = defaultdict(list)
+    for batch_idx, batch in enumerate(dl):
+        texts = batch["supervisions"]["text"]
+
+        hyps_dict = decode_one_batch(
+            params=params,
+            model=model,
+            sp=sp,
+            decoding_graph=decoding_graph,
+            batch=batch,
+        )
+
+        for name, hyps in hyps_dict.items():
+            this_batch = []
+            assert len(hyps) == len(texts)
+            for hyp_words, ref_text in zip(hyps, texts):
+                ref_words = ref_text.split()
+                this_batch.append((ref_words, hyp_words))
+
+            results[name].extend(this_batch)
+
+        num_cuts += len(texts)
+
+        if batch_idx % log_interval == 0:
+            batch_str = f"{batch_idx}/{num_batches}"
+
+            logging.info(
+                f"batch {batch_str}, cuts processed until now is {num_cuts}"
+            )
+    return results
+
+
+def save_results(
+    params: AttributeDict,
+    test_set_name: str,
+    results_dict: Dict[str, List[Tuple[List[int], List[int]]]],
+):
+    test_set_wers = dict()
+    test_set_cers = dict()
+    for key, results in results_dict.items():
+        recog_path = (
+            params.res_dir / f"recogs-{test_set_name}-{key}-{params.suffix}.txt"
+        )
+        store_transcripts(filename=recog_path, texts=results)
+        logging.info(f"The transcripts are stored in {recog_path}")
+
+        # The following prints out WERs, per-word error statistics and aligned
+        # ref/hyp pairs.
+        wers_filename = (
+            params.res_dir / f"wers-{test_set_name}-{key}-{params.suffix}.txt"
+        )
+        with open(wers_filename, "w") as f:
+            wer = write_error_stats(
+                f, f"{test_set_name}-{key}", results, enable_log=True
+            )
+            test_set_wers[key] = wer
+
+        # we also compute CER for spgispeech dataset.
+        results_char = []
+        for res in results:
+            results_char.append((list("".join(res[0])), list("".join(res[1]))))
+        cers_filename = (
+            params.res_dir / f"cers-{test_set_name}-{key}-{params.suffix}.txt"
+        )
+        with open(cers_filename, "w") as f:
+            cer = write_error_stats(
+                f, f"{test_set_name}-{key}", results_char, enable_log=True
+            )
+            test_set_cers[key] = cer
+
+        logging.info("Wrote detailed error stats to {}".format(wers_filename))
+
+    test_set_wers = {
+        k: v for k, v in sorted(test_set_wers.items(), key=lambda x: x[1])
+    }
+    test_set_cers = {
+        k: v for k, v in sorted(test_set_cers.items(), key=lambda x: x[1])
+    }
+    errs_info = (
+        params.res_dir
+        / f"wer-summary-{test_set_name}-{key}-{params.suffix}.txt"
+    )
+    with open(errs_info, "w") as f:
+        print("settings\tWER\tCER", file=f)
+        for key in test_set_wers:
+            print(
+                "{}\t{}\t{}".format(
+                    key, test_set_wers[key], test_set_cers[key]
+                ),
+                file=f,
+            )
+
+    s = "\nFor {}, WER/CER of different settings are:\n".format(test_set_name)
+    note = "\tbest for {}".format(test_set_name)
+    for key in test_set_wers:
+        s += "{}\t{}\t{}{}\n".format(
+            key, test_set_wers[key], test_set_cers[key], note
+        )
+        note = ""
+    logging.info(s)
+
+
+@torch.no_grad()
+def main():
+    parser = get_parser()
+    SPGISpeechAsrDataModule.add_arguments(parser)
+    args = parser.parse_args()
+    args.exp_dir = Path(args.exp_dir)
+
+    params = get_params()
+    params.update(vars(args))
+
+    assert params.decoding_method in (
+        "greedy_search",
+        "beam_search",
+        "fast_beam_search",
+        "modified_beam_search",
+    )
+    params.res_dir = params.exp_dir / params.decoding_method
+
+    if params.iter > 0:
+        params.suffix = f"iter-{params.iter}-avg-{params.avg}"
+    else:
+        params.suffix = f"epoch-{params.epoch}-avg-{params.avg}"
+
+    if "fast_beam_search" in params.decoding_method:
+        params.suffix += f"-beam-{params.beam}"
+        params.suffix += f"-max-contexts-{params.max_contexts}"
+        params.suffix += f"-max-states-{params.max_states}"
+    elif "beam_search" in params.decoding_method:
+        params.suffix += (
+            f"-{params.decoding_method}-beam-size-{params.beam_size}"
+        )
+    else:
+        params.suffix += f"-context-{params.context_size}"
+        params.suffix += f"-max-sym-per-frame-{params.max_sym_per_frame}"
+
+    setup_logger(f"{params.res_dir}/log-decode-{params.suffix}")
+    logging.info("Decoding started")
+
+    device = torch.device("cpu")
+    if torch.cuda.is_available():
+        device = torch.device("cuda", 0)
+
+    logging.info(f"Device: {device}")
+
+    sp = spm.SentencePieceProcessor()
+    sp.load(params.bpe_model)
+
+    # <blk> is defined in local/train_bpe_model.py
+    params.blank_id = sp.piece_to_id("<blk>")
+    params.vocab_size = sp.get_piece_size()
+
+    logging.info(params)
+
+    logging.info("About to create model")
+    model = get_transducer_model(params)
+
+    if params.iter > 0:
+        filenames = find_checkpoints(params.exp_dir, iteration=-params.iter)[
+            : params.avg
+        ]
+        if len(filenames) == 0:
+            raise ValueError(
+                f"No checkpoints found for"
+                f" --iter {params.iter}, --avg {params.avg}"
+            )
+        elif len(filenames) < params.avg:
+            raise ValueError(
+                f"Not enough checkpoints ({len(filenames)}) found for"
+                f" --iter {params.iter}, --avg {params.avg}"
+            )
+        logging.info(f"averaging {filenames}")
+        model.to(device)
+        model.load_state_dict(average_checkpoints(filenames, device=device))
+    elif params.avg == 1:
+        load_checkpoint(f"{params.exp_dir}/epoch-{params.epoch}.pt", model)
+    else:
+        start = params.epoch - params.avg + 1
+        filenames = []
+        for i in range(start, params.epoch + 1):
+            if start >= 0:
+                filenames.append(f"{params.exp_dir}/epoch-{i}.pt")
+        logging.info(f"averaging {filenames}")
+        model.to(device)
+        model.load_state_dict(average_checkpoints(filenames, device=device))
+
+    model.to(device)
+    model.eval()
+    model.device = device
+
+    if params.decoding_method == "fast_beam_search":
+        decoding_graph = k2.trivial_graph(params.vocab_size - 1, device=device)
+    else:
+        decoding_graph = None
+
+    num_param = sum([p.numel() for p in model.parameters()])
+    logging.info(f"Number of model parameters: {num_param}")
+
+    spgispeech = SPGISpeechAsrDataModule(args)
+
+    dev_cuts = spgispeech.dev_cuts()
+    val_cuts = spgispeech.val_cuts()
+
+    dev_dl = spgispeech.test_dataloaders(dev_cuts)
+    val_dl = spgispeech.test_dataloaders(val_cuts)
+
+    test_sets = ["dev", "val"]
+    test_dl = [dev_dl, val_dl]
+
+    for test_set, test_dl in zip(test_sets, test_dl):
+        results_dict = decode_dataset(
+            dl=test_dl,
+            params=params,
+            model=model,
+            sp=sp,
+            decoding_graph=decoding_graph,
+        )
+
+        save_results(
+            params=params,
+            test_set_name=test_set,
+            results_dict=results_dict,
+        )
+
+    logging.info("Done!")
+
+
+if __name__ == "__main__":
+    main()

egs/spgispeech/ASR/pruned_transducer_stateless2/decoder.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/decoder.py

egs/spgispeech/ASR/pruned_transducer_stateless2/encoder_interface.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/encoder_interface.py

egs/spgispeech/ASR/pruned_transducer_stateless2/export.py

@@ -0,0 +1,201 @@
+#!/usr/bin/env python3
+#
+# Copyright 2021 Xiaomi Corporation (Author: Fangjun Kuang)
+#
+# See ../../../../LICENSE for clarification regarding multiple authors
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+# This script converts several saved checkpoints
+# to a single one using model averaging.
+"""
+Usage:
+./pruned_transducer_stateless2/export.py \
+  --exp-dir ./pruned_transducer_stateless2/exp \
+  --bpe-model data/lang_bpe_500/bpe.model \
+  --avg-last-n 10
+
+It will generate a file exp_dir/pretrained.pt
+
+To use the generated file with `pruned_transducer_stateless2/decode.py`,
+you can do:
+
+    cd /path/to/exp_dir
+    ln -s pretrained.pt epoch-9999.pt
+
+    cd /path/to/egs/spgispeech/ASR
+    ./pruned_transducer_stateless2/decode.py \
+        --exp-dir ./pruned_transducer_stateless2/exp \
+        --epoch 9999 \
+        --avg 1 \
+        --max-duration 100 \
+        --bpe-model data/lang_bpe_500/bpe.model
+"""
+
+import argparse
+import logging
+from pathlib import Path
+
+import sentencepiece as spm
+import torch
+from train import get_params, get_transducer_model
+
+from icefall.checkpoint import (
+    average_checkpoints,
+    find_checkpoints,
+    load_checkpoint,
+)
+from icefall.utils import str2bool
+
+
+def get_parser():
+    parser = argparse.ArgumentParser(
+        formatter_class=argparse.ArgumentDefaultsHelpFormatter
+    )
+
+    parser.add_argument(
+        "--epoch",
+        type=int,
+        default=28,
+        help="It specifies the checkpoint to use for decoding."
+        "Note: Epoch counts from 0.",
+    )
+
+    parser.add_argument(
+        "--avg",
+        type=int,
+        default=15,
+        help="Number of checkpoints to average. Automatically select "
+        "consecutive checkpoints before the checkpoint specified by "
+        "'--epoch'. ",
+    )
+
+    parser.add_argument(
+        "--avg-last-n",
+        type=int,
+        default=0,
+        help="""If positive, --epoch and --avg are ignored and it
+        will use the last n checkpoints exp_dir/checkpoint-xxx.pt
+        where xxx is the number of processed batches while
+        saving that checkpoint.
+        """,
+    )
+
+    parser.add_argument(
+        "--exp-dir",
+        type=str,
+        default="pruned_transducer_stateless2/exp",
+        help="""It specifies the directory where all training related
+        files, e.g., checkpoints, log, etc, are saved
+        """,
+    )
+
+    parser.add_argument(
+        "--bpe-model",
+        type=str,
+        default="data/lang_bpe_500/bpe.model",
+        help="Path to the BPE model",
+    )
+
+    parser.add_argument(
+        "--jit",
+        type=str2bool,
+        default=False,
+        help="""True to save a model after applying torch.jit.script.
+        """,
+    )
+
+    parser.add_argument(
+        "--context-size",
+        type=int,
+        default=2,
+        help="The context size in the decoder. 1 means bigram; "
+        "2 means tri-gram",
+    )
+
+    return parser
+
+
+def main():
+    args = get_parser().parse_args()
+    args.exp_dir = Path(args.exp_dir)
+
+    assert args.jit is False, "Support torchscript will be added later"
+
+    params = get_params()
+    params.update(vars(args))
+
+    device = torch.device("cpu")
+    if torch.cuda.is_available():
+        device = torch.device("cuda", 0)
+
+    logging.info(f"device: {device}")
+
+    sp = spm.SentencePieceProcessor()
+    sp.load(params.bpe_model)
+
+    # <blk> is defined in local/train_bpe_model.py
+    params.blank_id = sp.piece_to_id("<blk>")
+    params.vocab_size = sp.get_piece_size()
+
+    logging.info(params)
+
+    logging.info("About to create model")
+    model = get_transducer_model(params)
+
+    model.to(device)
+
+    if params.avg_last_n > 0:
+        filenames = find_checkpoints(params.exp_dir)[: params.avg_last_n]
+        logging.info(f"averaging {filenames}")
+        model.to(device)
+        model.load_state_dict(average_checkpoints(filenames, device=device))
+    elif params.avg == 1:
+        load_checkpoint(f"{params.exp_dir}/epoch-{params.epoch}.pt", model)
+    else:
+        start = params.epoch - params.avg + 1
+        filenames = []
+        for i in range(start, params.epoch + 1):
+            if start >= 0:
+                filenames.append(f"{params.exp_dir}/epoch-{i}.pt")
+        logging.info(f"averaging {filenames}")
+        model.to(device)
+        model.load_state_dict(average_checkpoints(filenames, device=device))
+
+    model.eval()
+
+    model.to("cpu")
+    model.eval()
+
+    if params.jit:
+        logging.info("Using torch.jit.script")
+        model = torch.jit.script(model)
+        filename = params.exp_dir / "cpu_jit.pt"
+        model.save(str(filename))
+        logging.info(f"Saved to {filename}")
+    else:
+        logging.info("Not using torch.jit.script")
+        # Save it using a format so that it can be loaded
+        # by :func:`load_checkpoint`
+        filename = params.exp_dir / "pretrained.pt"
+        torch.save({"model": model.state_dict()}, str(filename))
+        logging.info(f"Saved to {filename}")
+
+
+if __name__ == "__main__":
+    formatter = (
+        "%(asctime)s %(levelname)s [%(filename)s:%(lineno)d] %(message)s"
+    )
+
+    logging.basicConfig(format=formatter, level=logging.INFO)
+    main()

egs/spgispeech/ASR/pruned_transducer_stateless2/joiner.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/joiner.py

egs/spgispeech/ASR/pruned_transducer_stateless2/model.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/model.py

egs/spgispeech/ASR/pruned_transducer_stateless2/optim.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/optim.py

egs/spgispeech/ASR/pruned_transducer_stateless2/scaling.py

@@ -0,0 +1 @@
+../../../librispeech/ASR/pruned_transducer_stateless2/scaling.py

egs/spgispeech/ASR/shared

@@ -0,0 +1 @@
+../../../icefall/shared/

egs/yesno/ASR/local/compute_fbank_yesno.py

@@ -38,7 +38,9 @@ def compute_fbank_yesno():
         "test",
     )
     manifests = read_manifests_if_cached(
-        dataset_parts=dataset_parts, output_dir=src_dir
+        dataset_parts=dataset_parts,
+        output_dir=src_dir,
+        prefix="yesno",
     )
     assert manifests is not None
 

icefall/diagnostics.py

@@ -18,7 +18,8 @@
 
 
 import random
-from typing import List, Optional, Tuple
+from dataclasses import dataclass
+from typing import Optional, Tuple
 
 import torch
 from torch import Tensor, nn
@@ -28,16 +29,12 @@ class TensorDiagnosticOptions(object):
     """Options object for tensor diagnostics:
 
     Args:
-      memory_limit:
-        The maximum number of bytes per tensor
-        (limits how many copies of the tensor we cache).
       max_eig_dim:
         The maximum dimension for which we print out eigenvalues
         (limited for speed reasons).
     """
 
-    def __init__(self, memory_limit: int = (2 ** 20), max_eig_dim: int = 512):
+    def __init__(self, max_eig_dim: int = 512):
-        self.memory_limit = memory_limit
         self.max_eig_dim = max_eig_dim
 
     def dim_is_summarized(self, size: int):
@@ -94,138 +91,10 @@ def get_tensor_stats(
     return x, count
 
 
-def get_diagnostics_for_dim(
+@dataclass
-    dim: int,
+class TensorAndCount:
-    tensors: List[Tensor],
+    tensor: Tensor
-    options: TensorDiagnosticOptions,
+    count: int
-    sizes_same: bool,
-    stats_type: str,
-) -> str:
-    """
-    This function gets diagnostics for a dimension of a module.
-
-    Args:
-      dim:
-        the dimension to analyze, with 0 <= dim < tensors[0].ndim
-      options:
-        options object
-      sizes_same:
-        True if all the tensor sizes are the same on this dimension
-      stats_type: either "abs" or "positive" or "eigs" or "value",
-        imdictates the type of stats we accumulate, abs is mean absolute
-        value, "positive" is proportion of positive to nonnegative values,
-        "eigs" is eigenvalues after doing outer product on this dim, sum
-        over all other dimes.
-    Returns:
-      Diagnostic as a string, either percentiles or the actual values,
-      see the code.  Will return the empty string if the diagnostics did
-      not make sense to print out for this dimension, e.g. dimension
-      mismatch and stats_type == "eigs".
-    """
-
-    # stats_and_counts is a list of pair (Tensor, int)
-    stats_and_counts = [get_tensor_stats(x, dim, stats_type) for x in tensors]
-    stats = [x[0] for x in stats_and_counts]
-    counts = [x[1] for x in stats_and_counts]
-
-    if stats_type == "eigs":
-        try:
-            stats = torch.stack(stats).sum(dim=0)
-        except:  # noqa
-            return ""
-        count = sum(counts)
-        stats = stats / count
-        try:
-            eigs, _ = torch.symeig(stats)
-            stats = eigs.abs().sqrt()
-        except:  # noqa
-            print("Error getting eigenvalues, trying another method.")
-            eigs = torch.linalg.eigvals(stats)
-            stats = eigs.abs().sqrt()
-        # sqrt so it reflects data magnitude, like stddev- not variance
-    elif sizes_same:
-        stats = torch.stack(stats).sum(dim=0)
-        count = sum(counts)
-        stats = stats / count
-    else:
-        stats = [x[0] / x[1] for x in stats_and_counts]
-        stats = torch.cat(stats, dim=0)
-    if stats_type == "rms":
-        stats = stats.sqrt()
-
-    # if `summarize` we print percentiles of the stats; else,
-    # we print out individual elements.
-    summarize = (not sizes_same) or options.dim_is_summarized(stats.numel())
-    if summarize:
-        # print out percentiles.
-        stats = stats.sort()[0]
-        num_percentiles = 10
-        size = stats.numel()
-        percentiles = []
-        for i in range(num_percentiles + 1):
-            index = (i * (size - 1)) // num_percentiles
-            percentiles.append(stats[index].item())
-        percentiles = ["%.2g" % x for x in percentiles]
-        percentiles = " ".join(percentiles)
-        ans = f"percentiles: [{percentiles}]"
-    else:
-        ans = stats.tolist()
-        ans = ["%.2g" % x for x in ans]
-        ans = "[" + " ".join(ans) + "]"
-    if stats_type == "value":
-        # This norm is useful because it is strictly less than the largest
-        # sqrt(eigenvalue) of the variance, which we print out, and shows,
-        # speaking in an approximate way, how much of that largest eigenvalue
-        # can be attributed to the mean of the distribution.
-        norm = (stats ** 2).sum().sqrt().item()
-        mean = stats.mean().item()
-        rms = (stats ** 2).mean().sqrt().item()
-        ans += f", norm={norm:.2g}, mean={mean:.2g}, rms={rms:.2g}"
-    else:
-        mean = stats.mean().item()
-        rms = (stats ** 2).mean().sqrt().item()
-        ans += f", mean={mean:.2g}, rms={rms:.2g}"
-    return ans
-
-
-def print_diagnostics_for_dim(
-    name: str, dim: int, tensors: List[Tensor], options: TensorDiagnosticOptions
-):
-    """This function prints diagnostics for a dimension of a tensor.
-
-    Args:
-      name:
-        The tensor name.
-      dim:
-        The dimension to analyze, with 0 <= dim < tensors[0].ndim.
-      tensors:
-        List of cached tensors to get the stats.
-      options:
-        Options object.
-    """
-
-    ndim = tensors[0].ndim
-    if ndim > 1:
-        stats_types = ["abs", "positive", "value", "rms"]
-        if tensors[0].shape[dim] <= options.max_eig_dim:
-            stats_types.append("eigs")
-    else:
-        stats_types = ["value", "abs"]
-
-    for stats_type in stats_types:
-        sizes = [x.shape[dim] for x in tensors]
-        sizes_same = all([x == sizes[0] for x in sizes])
-        s = get_diagnostics_for_dim(
-            dim, tensors, options, sizes_same, stats_type
-        )
-        if s == "":
-            continue
-
-        min_size = min(sizes)
-        max_size = max(sizes)
-        size_str = f"{min_size}" if sizes_same else f"{min_size}..{max_size}"
-        # stats_type will be "abs" or "positive".
-        print(f"module={name}, dim={dim}, size={size_str}, {stats_type} {s}")
 
 
 class TensorDiagnostic(object):
@@ -242,8 +111,18 @@ class TensorDiagnostic(object):
     def __init__(self, opts: TensorDiagnosticOptions, name: str):
         self.name = name
         self.opts = opts
-        # A list to cache the tensors.
+
-        self.saved_tensors = []
+        self.stats = None  # we'll later assign a list to this data member.  It's a list of dict.
+
+        # the keys into self.stats[dim] are strings, whose values can be
+        # "abs", "value", "positive", "rms", "value".
+        # The values e.g. self.stats[dim]["rms"] are lists of dataclass TensorAndCount,
+        # containing a tensor and its associated count (which is the sum of the other dims
+        # that we aggregated over, e.g. the number of frames and/or batch elements and/or
+        # channels.
+        # ... we actually accumulate the Tensors / counts any time we have the same-dim tensor,
+        # only adding a new element to the list if there was a different dim.
+        # if the string in the key is "eigs", if we detect a length mismatch we put None as the value.
 
     def accumulate(self, x):
         """Accumulate tensors."""
@@ -251,50 +130,130 @@ class TensorDiagnostic(object):
             x = x[0]
         if not isinstance(x, Tensor):
             return
-        if x.device == torch.device("cpu"):
+        x = x.detach().clone()
-            x = x.detach().clone()
+        if x.ndim == 0:
-        else:
+            x = x.unsqueeze(0)
-            x = x.detach().to("cpu", non_blocking=True)
+        ndim = x.ndim
-        self.saved_tensors.append(x)
+        if self.stats is None:
-        num = len(self.saved_tensors)
+            self.stats = [dict() for _ in range(ndim)]
-        if num & (num - 1) == 0:  # power of 2..
-            self._limit_memory()
 
-    def _limit_memory(self):
+        for dim in range(ndim):
-        """Only keep the newly cached tensors to limit memory."""
+            this_dim_stats = self.stats[dim]
-        if len(self.saved_tensors) > 1024:
+            if ndim > 1:
-            self.saved_tensors = self.saved_tensors[-1024:]
+                stats_types = ["abs", "positive", "value", "rms"]
-            return
+                if x.shape[dim] <= self.opts.max_eig_dim:
+                    stats_types.append("eigs")
+            else:
+                stats_types = ["value", "abs"]
 
-        tot_mem = 0.0
+            for stats_type in stats_types:
-        for i in reversed(range(len(self.saved_tensors))):
+                stats, count = get_tensor_stats(x, dim, stats_type)
-            tot_mem += (
+                if stats_type not in this_dim_stats:
-                self.saved_tensors[i].numel()
+                    this_dim_stats[stats_type] = []  # list of TensorAndCount
-                * self.saved_tensors[i].element_size()
+
-            )
+                done = False
-            if tot_mem > self.opts.memory_limit:
+                if this_dim_stats[stats_type] is None:
-                self.saved_tensors = self.saved_tensors[i:]
+                    # we can reach here if we detected for stats_type "eigs" that
-                return
+                    # where was more than one different size for this dim.  Then we
+                    # disable accumulating this stats type, as it uses too much memory.
+                    continue
+                for s in this_dim_stats[stats_type]:
+                    if s.tensor.shape == stats.shape:
+                        s.tensor += stats
+                        s.count += count
+                        done = True
+                        break
+                if not done:
+                    if (
+                        this_dim_stats[stats_type] != []
+                        and stats_type == "eigs"
+                    ):
+                        # >1 size encountered on this dim, e.g. it's a batch or time dimension,
+                        # don't accumulat "eigs" stats type, it uses too much memory
+                        this_dim_stats[stats_type] = None
+                    else:
+                        this_dim_stats[stats_type].append(
+                            TensorAndCount(stats, count)
+                        )
 
     def print_diagnostics(self):
         """Print diagnostics for each dimension of the tensor."""
-        if len(self.saved_tensors) == 0:
+        for dim, this_dim_stats in enumerate(self.stats):
-            print("{name}: no stats".format(name=self.name))
+            for stats_type, stats_list in this_dim_stats.items():
-            return
+                # stats_type could be "rms", "value", "abs", "eigs", "positive".
+                # "value" could be a list of TensorAndCount, or None
+                if stats_list is None:
+                    assert stats_type == "eigs"
+                    continue
 
-        if self.saved_tensors[0].ndim == 0:
+                if stats_type == "eigs":
-            # Ensure there is at least one dim.
+                    assert len(stats_list) == 1
-            self.saved_tensors = [x.unsqueeze(0) for x in self.saved_tensors]
+                    stats = stats_list[0].tensor / stats_list[0].count
+                    try:
+                        eigs, _ = torch.symeig(stats)
+                        stats = eigs.abs().sqrt()
+                    except:  # noqa
+                        print(
+                            "Error getting eigenvalues, trying another method."
+                        )
+                        eigs = torch.linalg.eigvals(stats)
+                        stats = eigs.abs().sqrt()
+                        # sqrt so it reflects data magnitude, like stddev- not variance
+                elif len(stats_list) == 1:
+                    stats = stats_list[0].tensor / stats_list[0].count
+                else:
+                    stats = torch.cat(
+                        [x.tensor / x.count for x in stats_list], dim=0
+                    )
 
-        try:
+                if stats_type == "rms":
-            device = torch.device("cuda")
+                    # we stored the square; after aggregation we need to take sqrt.
-        except:  # noqa
+                    stats = stats.sqrt()
-            device = torch.device("cpu")
 
-        ndim = self.saved_tensors[0].ndim
+                # if `summarize` we print percentiles of the stats; else,
-        tensors = [x.to(device) for x in self.saved_tensors]
+                # we print out individual elements.
-        for dim in range(ndim):
+                summarize = (
-            print_diagnostics_for_dim(self.name, dim, tensors, self.opts)
+                    len(stats_list) > 1
+                ) or self.opts.dim_is_summarized(stats.numel())
+                if summarize:  # usually `summarize` will be true
+                    # print out percentiles.
+                    stats = stats.sort()[0]
+                    num_percentiles = 10
+                    size = stats.numel()
+                    percentiles = []
+                    for i in range(num_percentiles + 1):
+                        index = (i * (size - 1)) // num_percentiles
+                        percentiles.append(stats[index].item())
+                    percentiles = ["%.2g" % x for x in percentiles]
+                    percentiles = " ".join(percentiles)
+                    ans = f"percentiles: [{percentiles}]"
+                else:
+                    ans = stats.tolist()
+                    ans = ["%.2g" % x for x in ans]
+                    ans = "[" + " ".join(ans) + "]"
+                if stats_type == "value":
+                    # This norm is useful because it is strictly less than the largest
+                    # sqrt(eigenvalue) of the variance, which we print out, and shows,
+                    # speaking in an approximate way, how much of that largest eigenvalue
+                    # can be attributed to the mean of the distribution.
+                    norm = (stats ** 2).sum().sqrt().item()
+                    ans += f", norm={norm:.2g}"
+                mean = stats.mean().item()
+                rms = (stats ** 2).mean().sqrt().item()
+                ans += f", mean={mean:.2g}, rms={rms:.2g}"
+
+                # OK, "ans" contains the actual stats, e.g.
+                # ans = "percentiles: [0.43 0.46 0.48 0.49 0.49 0.5 0.51 0.52 0.53 0.54 0.59], mean=0.5, rms=0.5"
+
+                sizes = [x.tensor.shape[0] for x in stats_list]
+                size_str = (
+                    f"{sizes[0]}"
+                    if len(sizes) == 1
+                    else f"{min(sizes)}..{max(sizes)}"
+                )
+                print(
+                    f"module={self.name}, dim={dim}, size={size_str}, {stats_type} {ans}"
+                )
 
 
 class ModelDiagnostic(object):
@@ -326,7 +285,7 @@ class ModelDiagnostic(object):
 
 
 def attach_diagnostics(
-    model: nn.Module, opts: TensorDiagnosticOptions
+    model: nn.Module, opts: Optional[TensorDiagnosticOptions] = None
 ) -> ModelDiagnostic:
     """Attach a ModelDiagnostic object to the model by
     1) registering forward hook and backward hook on each module, to accumulate
@@ -389,7 +348,7 @@ def attach_diagnostics(
 
 
 def _test_tensor_diagnostic():
-    opts = TensorDiagnosticOptions(2 ** 20, 512)
+    opts = TensorDiagnosticOptions(512)
 
     diagnostic = TensorDiagnostic(opts, "foo")