mirrors/icefall
1
0
Fork 0
mirror of https://github.com/k2-fsa/icefall.git synced 2025-09-08 08:34:19 +00:00
Code Issues Packages Projects Releases Wiki Activity

minor fixes

Browse Source
This commit is contained in:
JinZr 2023-09-02 01:04:50 +08:00
parent cec73bd28b
commit aadb4507bf
2 changed files with 14 additions and 18 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

8
egs/multi_zh-hans/ASR/zipformer/export.py
View File

@ -144,17 +144,13 @@ Note: If you don't want to train a model from scratch, we have
provided one for you. You can get it at
- non-streaming model:
https://huggingface.co/Zengwei/icefall-asr-librispeech-zipformer-2023-05-15
- streaming model:
https://huggingface.co/Zengwei/icefall-asr-librispeech-streaming-zipformer-2023-05-17
https://huggingface.co/zrjin/icefall-asr-multi-zh-hans-zipformer-2023-9-2/
with the following commands:
sudo apt-get install git-lfs
git lfs install
git clone https://huggingface.co/Zengwei/icefall-asr-librispeech-zipformer-2023-05-15
git clone https://huggingface.co/Zengwei/icefall-asr-librispeech-streaming-zipformer-2023-05-17
git clone https://huggingface.co/zrjin/icefall-asr-multi-zh-hans-zipformer-2023-9-2/
# You will find the pre-trained models in exp dir
"""

24
egs/multi_zh-hans/ASR/zipformer/pretrained.py
View File

@ -25,18 +25,18 @@ dataset, you should change the argument values according to your dataset.
./zipformer/export.py \
--exp-dir ./zipformer/exp \
--tokens data/lang_bpe_500/tokens.txt \
--epoch 30 \
--avg 9
--tokens data/lang_bpe_2000/tokens.txt \
--epoch 23 \
--avg 1
- For streaming model:
./zipformer/export.py \
--exp-dir ./zipformer/exp \
--causal 1 \
--tokens data/lang_bpe_500/tokens.txt \
--epoch 30 \
--avg 9
--tokens data/lang_bpe_2000/tokens.txt \
--epoch 23 \
--avg 1
Usage of this script:
@ -45,7 +45,7 @@ Usage of this script:
(1) greedy search
./zipformer/pretrained.py \
--checkpoint ./zipformer/exp/pretrained.pt \
--tokens data/lang_bpe_500/tokens.txt \
--tokens data/lang_bpe_2000/tokens.txt \
--method greedy_search \
/path/to/foo.wav \
/path/to/bar.wav
@ -53,7 +53,7 @@ Usage of this script:
(2) modified beam search
./zipformer/pretrained.py \
--checkpoint ./zipformer/exp/pretrained.pt \
--tokens ./data/lang_bpe_500/tokens.txt \
--tokens ./data/lang_bpe_2000/tokens.txt \
--method modified_beam_search \
/path/to/foo.wav \
/path/to/bar.wav
@ -61,7 +61,7 @@ Usage of this script:
(3) fast beam search
./zipformer/pretrained.py \
--checkpoint ./zipformer/exp/pretrained.pt \
--tokens ./data/lang_bpe_500/tokens.txt \
--tokens ./data/lang_bpe_2000/tokens.txt \
--method fast_beam_search \
/path/to/foo.wav \
/path/to/bar.wav
@ -74,7 +74,7 @@ Usage of this script:
--causal 1 \
--chunk-size 16 \
--left-context-frames 128 \
--tokens ./data/lang_bpe_500/tokens.txt \
--tokens ./data/lang_bpe_2000/tokens.txt \
--method greedy_search \
/path/to/foo.wav \
/path/to/bar.wav
@ -85,7 +85,7 @@ Usage of this script:
--causal 1 \
--chunk-size 16 \
--left-context-frames 128 \
--tokens ./data/lang_bpe_500/tokens.txt \
--tokens ./data/lang_bpe_2000/tokens.txt \
--method modified_beam_search \
/path/to/foo.wav \
/path/to/bar.wav
@ -96,7 +96,7 @@ Usage of this script:
--causal 1 \
--chunk-size 16 \
--left-context-frames 128 \
--tokens ./data/lang_bpe_500/tokens.txt \
--tokens ./data/lang_bpe_2000/tokens.txt \
--method fast_beam_search \
/path/to/foo.wav \
/path/to/bar.wav