mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-09-19 05:54:20 +00:00
Code Issues Packages Projects Releases Wiki Activity

minor fixes

Browse Source
This commit is contained in:
marcoyang 2023-06-29 12:33:26 +08:00
parent 2f1af8f303
commit a6462490fb
2 changed files with 17 additions and 13 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
docs/source/decoding-with-langugage-models/LODR.rst
View File

@ -70,9 +70,9 @@ As the initial step, let's download the pre-trained model.
.. code-block:: bash .. code-block:: bash
$ git lfs install $ GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/Zengwei/icefall-asr-librispeech-pruned-transducer-stateless7-streaming-2022-12-29
$ git clone https://huggingface.co/Zengwei/icefall-asr-librispeech-pruned-transducer-stateless7-streaming-2022-12-29
$ pushd icefall-asr-librispeech-pruned-transducer-stateless7-streaming-2022-12-29/exp $ pushd icefall-asr-librispeech-pruned-transducer-stateless7-streaming-2022-12-29/exp
$ git lfs pull --include "pretrained.pt"
$ ln -s pretrained.pt epoch-99.pt # create a symbolic link so that the checkpoint can be loaded $ ln -s pretrained.pt epoch-99.pt # create a symbolic link so that the checkpoint can be loaded
To test the model, let's have a look at the decoding results **without** using LM. This can be done via the following command: To test the model, let's have a look at the decoding results **without** using LM. This can be done via the following command:
@ -81,8 +81,9 @@ To test the model, let's have a look at the decoding results **without** using L
$ exp_dir=./icefall-asr-librispeech-pruned-transducer-stateless7-streaming-2022-12-29/exp/ $ exp_dir=./icefall-asr-librispeech-pruned-transducer-stateless7-streaming-2022-12-29/exp/
$ ./pruned_transducer_stateless7_streaming/decode.py \ $ ./pruned_transducer_stateless7_streaming/decode.py \
--epoch 30 \ --epoch 99 \
--avg 9 \ --avg 1 \
--use-averaged-model False \
--exp-dir $exp_dir \ --exp-dir $exp_dir \
--max-duration 600 \ --max-duration 600 \
--decode-chunk-len 32 \ --decode-chunk-len 32 \
@ -102,11 +103,11 @@ Note that the bi-gram is estimated on the LibriSpeech 960 hours' text.
.. code-block:: bash .. code-block:: bash
$ git lfs install
$ # download the external LM $ # download the external LM
$ git clone https://huggingface.co/ezerhouni/icefall-librispeech-rnn-lm $ GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/ezerhouni/icefall-librispeech-rnn-lm
$ # create a symbolic link so that the checkpoint can be loaded $ # create a symbolic link so that the checkpoint can be loaded
$ pushd icefall-librispeech-rnn-lm/exp $ pushd icefall-librispeech-rnn-lm/exp
$ git lfs pull --include "pretrained.pt"
$ ln -s pretrained.pt epoch-99.pt $ ln -s pretrained.pt epoch-99.pt
$ popd $ popd
$ $

17
docs/source/decoding-with-langugage-models/shallow-fusion.rst
View File

@ -31,9 +31,9 @@ As the initial step, let's download the pre-trained model.
.. code-block:: bash .. code-block:: bash
$ git lfs install $ GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/Zengwei/icefall-asr-librispeech-pruned-transducer-stateless7-streaming-2022-12-29
$ git clone https://huggingface.co/Zengwei/icefall-asr-librispeech-pruned-transducer-stateless7-streaming-2022-12-29
$ pushd icefall-asr-librispeech-pruned-transducer-stateless7-streaming-2022-12-29/exp $ pushd icefall-asr-librispeech-pruned-transducer-stateless7-streaming-2022-12-29/exp
$ git lfs pull --include "pretrained.pt"
$ ln -s pretrained.pt epoch-99.pt # create a symbolic link so that the checkpoint can be loaded $ ln -s pretrained.pt epoch-99.pt # create a symbolic link so that the checkpoint can be loaded
To test the model, let's have a look at the decoding results without using LM. This can be done via the following command: To test the model, let's have a look at the decoding results without using LM. This can be done via the following command:
@ -42,8 +42,9 @@ To test the model, let's have a look at the decoding results without using LM. T
$ exp_dir=./icefall-asr-librispeech-pruned-transducer-stateless7-streaming-2022-12-29/exp/ $ exp_dir=./icefall-asr-librispeech-pruned-transducer-stateless7-streaming-2022-12-29/exp/
$ ./pruned_transducer_stateless7_streaming/decode.py \ $ ./pruned_transducer_stateless7_streaming/decode.py \
--epoch 30 \ --epoch 99 \
--avg 9 \ --avg 1 \
--use-averaged-model False \
--exp-dir $exp_dir \ --exp-dir $exp_dir \
--max-duration 600 \ --max-duration 600 \
--decode-chunk-len 32 \ --decode-chunk-len 32 \
@ -63,10 +64,12 @@ Training a language model usually takes a long time, we can download a pre-train
.. code-block:: bash .. code-block:: bash
$ git lfs install $ # download the external LM
$ git clone https://huggingface.co/ezerhouni/icefall-librispeech-rnn-lm $ GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/ezerhouni/icefall-librispeech-rnn-lm
$ # create a symbolic link so that the checkpoint can be loaded
$ pushd icefall-librispeech-rnn-lm/exp $ pushd icefall-librispeech-rnn-lm/exp
$ ln -s pretrained.pt epoch-99.pt # create a symbolic link so that the checkpoint can be loaded $ git lfs pull --include "pretrained.pt"
$ ln -s pretrained.pt epoch-99.pt
$ popd $ popd
.. note:: .. note::