804 Commits

Author SHA1 Message Date
Fangjun Kuang
2318c3fbd0
Support CTC decoding on CPU using OpenFst and kaldi decoders. (#1244) 2023-09-26 16:36:19 +08:00
zr_jin
1b565dd251
added softlinks to local dir (#1273) 2023-09-26 15:41:39 +08:00
marcoyang1998
e17f884ace
Fix docs for MVQ (#1272)
* typo fix
2023-09-25 15:36:40 +08:00
zr_jin
ef5da4824d
formatted the entire LibriSpeech recipe (#1270)
* formatted the entire librispeech recipe

* minor updates
2023-09-24 17:31:01 +08:00
zr_jin
ef658d691e
fixes for init value of diagnostics.TensorDiagnosticOptions (#1269)
* fixes for `diagnostics`

Replace `2 ** 22` with `512` as the default value of `diagnostics.TensorDiagnosticOptions`

also black formatted some scripts

* fixed formatting issues
2023-09-24 17:06:47 +08:00
Fangjun Kuang
34e40a86b3
Fix exporting decoder model to onnx (#1264)
* Use torch.jit.script() to export the decoder model

See also https://github.com/k2-fsa/sherpa-onnx/issues/327
2023-09-22 09:57:15 +08:00
Fangjun Kuang
f5dc957d44
Fix CI tests (#1266) 2023-09-21 21:16:14 +08:00
l2009312042
45d60ef262
Update conformer.py (#1200)
* Update conformer.py
* Update zipformer.py

fix bug in get_dynamic_dropout_rate
2023-09-21 19:41:10 +08:00
zr_jin
bbb03f7962
Update decoder.py (#1262) 2023-09-20 08:15:54 +08:00
Tiance Wang
7e1288af50
fix thchs-30 download command (#1260) 2023-09-19 16:46:36 +08:00
zr_jin
565d2c2f5b
Minor fixes to the libricss recipe (#1256) 2023-09-15 02:37:53 +08:00
docterstrange
fba1710622
modify tal_csasr recipe (#1252)
Co-authored-by: zss11 <zss11@d3-hpc-sjtu-test-001.cm.cluster>
2023-09-14 09:58:28 +08:00
zr_jin
7cc2dae940
Fixes to incorporate with the latest Lhotse release (#1249) 2023-09-13 12:39:49 +08:00
zr_jin
0f1bc6f8af
Multi_zh-Hans Recipe (#1238)
* Init commit for recipes trained on multiple zh datasets.

* fbank extraction for thchs30

* added support for aishell1

* added support for aishell-2

* fixes

* fixes

* fixes

* added support for stcmds and primewords

* fixes

* added support for magicdata

script for fbank computation not done yet

* added script for magicdata fbank computation

* file permission fixed

* updated for the wenetspeech recipe

* updated

* Update preprocess_kespeech.py

* updated

* updated

* updated

* updated

* file permission fixed

* updated paths

* fixes

* added support for kespeech dev/test set fbank computation

* fixes for file permission

* refined support for KeSpeech

* added scripts for BPE model training

* updated

* init commit for the multi_zh-cn zipformer recipe

* disable speed perturbation by default

* updated

* updated

* added necessary files for the zipformer recipe

* removed redundant wenetspeech M and S sets

* updates for multi dataset decoding

* refined

* formatting issues fixed

* updated

* minor fixes

* this commit finalize the recipe (hopefully)

* fixed formatting issues

* minor fixes

* updated

* using soft links to reduce redundancy

* minor updates

* using soft links to reduce redundancy

* minor updates

* minor updates

* using soft links to reduce redundancy

* minor updates

* Update README.md

* minor updates

* Update egs/multi_zh-hans/ASR/local/compute_fbank_magicdata.py

Co-authored-by: Fangjun Kuang <csukuangfj@gmail.com>

* Update egs/multi_zh-hans/ASR/local/compute_fbank_magicdata.py

Co-authored-by: Fangjun Kuang <csukuangfj@gmail.com>

* Update egs/multi_zh-hans/ASR/local/compute_fbank_stcmds.py

Co-authored-by: Fangjun Kuang <csukuangfj@gmail.com>

* Update egs/multi_zh-hans/ASR/local/compute_fbank_stcmds.py

Co-authored-by: Fangjun Kuang <csukuangfj@gmail.com>

* Update egs/multi_zh-hans/ASR/local/compute_fbank_primewords.py

Co-authored-by: Fangjun Kuang <csukuangfj@gmail.com>

* Update egs/multi_zh-hans/ASR/local/compute_fbank_primewords.py

Co-authored-by: Fangjun Kuang <csukuangfj@gmail.com>

* minor updates

* minor fixes

* fixed a formatting issue

* Update preprocess_kespeech.py

* Update prepare.sh

* Update egs/multi_zh-hans/ASR/local/compute_fbank_kespeech_splits.py

Co-authored-by: Fangjun Kuang <csukuangfj@gmail.com>

* Update egs/multi_zh-hans/ASR/local/preprocess_kespeech.py

Co-authored-by: Fangjun Kuang <csukuangfj@gmail.com>

* removed redundant files

* symlinks added

* minor updates

* added CI tests for `multi_zh-hans`

* minor fixes

* Update run-multi-zh_hans-zipformer.sh

* Update run-multi-zh_hans-zipformer.sh

* Update run-multi-zh_hans-zipformer.sh

* Update run-multi-zh_hans-zipformer.sh

* Update run-multi-zh_hans-zipformer.sh

* Update run-multi-zh_hans-zipformer.sh

* Update run-multi-zh_hans-zipformer.sh

---------

Co-authored-by: Fangjun Kuang <csukuangfj@gmail.com>
2023-09-13 11:57:05 +08:00
zr_jin
d50a9ea030
doc str fixes (#1241) 2023-09-07 16:34:53 +08:00
zr_jin
9ef8145fa3
minor fixes (#1240) 2023-09-04 17:56:05 +08:00
Desh Raj
8fcadb68a7
Missing definitions in scaling.py added (#1232) 2023-08-31 10:31:05 +08:00
Wei Kang
4d7f73ce65
Add context biasing for zipformer recipe (#1204)
* Add context biasing for zipformer recipe

* support context biasing in modified_beam_search_LODR

* fix context graph

* Minor fixes
2023-08-28 19:37:32 +08:00
Fangjun Kuang
fc2df07841
Add icefall tutorials for dummies. (#1220) 2023-08-16 22:32:41 +08:00
Erwan Zerhouni
9a47c08d08
Update padding modified beam search (#1217) 2023-08-14 16:10:50 +02:00
Piotr Żelasko
b0e8a40c89
Speed up yesno training to finish in ~10s on CPU (#1215) 2023-08-13 09:50:59 +08:00
Fangjun Kuang
dfccadc6b6
Fix a typo in export_onnx.py for yesno (#1213) 2023-08-12 16:59:06 +08:00
zr_jin
a81396b482
Use tokens.txt to replace bpe.model (#1162) 2023-08-12 16:53:59 +08:00
Fangjun Kuang
d6b28a11a7
Add export script for the yesno recipe. (#1212) 2023-08-11 23:57:00 +08:00
zr_jin
74806b744b
disable speed perturbation by default (#1176)
* disable speed perturbation by default

* minor fixes

* minor updates

* updated bash scripts to incorporate with the `speed-perturb` arg

* minor fixes

1. changed the naming scheme from `speed-perturb` to `perturb-speed` to align with the librispeech recipe

>> 00256a7669/egs/librispeech/ASR/local/compute_fbank_librispeech.py (L65)

2. changed arg type for `perturb-speed` to str2bool
2023-08-10 20:56:02 +08:00
Yifan Yang
00256a7669
Fix decode_stream.py (#1208)
* FIx decode_stream.py

* Update decode_stream.py
2023-08-09 09:40:58 +08:00
marcoyang1998
1ee251c8b3
Decode zipformer with external LMs (#1193)
* update some documentation

* support decoding with LMs in zipformer recipe

* update RESULTS.md
2023-08-03 15:50:35 +08:00
kobenaxie
80d922c158
Update preprocess_commonvoice.py to fix text normalization bug. (#1181) 2023-07-26 16:54:42 +08:00
Fangjun Kuang
1dbbd7759e
Add tests for subsample.py and fix typos (#1180) 2023-07-25 14:46:18 +08:00
zr_jin
4ab7d61008
removed batch_name to fix a KeyError with "uttid" (#1172) 2023-07-15 12:39:32 +08:00
marcoyang1998
5ed6fc0e6d
add sym link (#1170) 2023-07-12 15:37:14 +08:00
Desh Raj
41b16d7838
SURT recipe for AMI and ICSI (#1133)
* merge upstream

* add SURT model and training

* add libricss decoding

* add chunk width randomization

* decode SURT with libricss

* initial commit for zipformer_ctc

* remove unwanted changes

* remove changes to other recipe

* fix zipformer softlink

* fix for JIT export

* add missing file

* fix symbolic links

* update results

* clean commit for SURT recipe

* training libricss surt model

* remove unwanted files

* remove unwanted changes

* remove changes in librispeech

* change some files to symlinks

* remove unwanted changes in utils

* add export script

* add README

* minor fix in README

* add assets for README

* replace some files with symlinks

* remove unused decoding methods

* initial commit for SURT AMI recipe

* fix symlink

* add train + decode scripts

* add missing symlink

* change files to symlink

* change file type
2023-07-08 23:01:51 +08:00
Yifan Yang
ffe816e2a8
Fix blank skip ci test (#1167)
* Fix for ci

* Fix frame_reducer
2023-07-06 23:12:41 +08:00
Fangjun Kuang
130ad0319d
Fix CI test for zipformer CTC (#1165) 2023-07-05 10:38:29 +08:00
Fangjun Kuang
b8a17944e4
Fix zipformer CI test (#1164) 2023-07-05 10:23:35 +08:00
Desh Raj
a4402b88e6
SURT multi-talker ASR recipe (#1126)
* merge upstream

* add SURT model and training

* add libricss decoding

* add chunk width randomization

* decode SURT with libricss

* initial commit for zipformer_ctc

* remove unwanted changes

* remove changes to other recipe

* fix zipformer softlink

* fix for JIT export

* add missing file

* fix symbolic links

* update results

* clean commit for SURT recipe

* training libricss surt model

* remove unwanted files

* remove unwanted changes

* remove changes in librispeech

* change some files to symlinks

* remove unwanted changes in utils

* add export script

* add README

* minor fix in README

* add assets for README

* replace some files with symlinks

* remove unused decoding methods

* fix symlink

* address comments from @csukuangfj
2023-07-04 19:25:58 +08:00
zr_jin
856c0f2a60
fixed default param for an aishell recipe (#1159) 2023-07-04 19:12:39 +08:00
Fangjun Kuang
9009d028a0
Fix ONNX export for the latest non-streaming zipformer. (#1160) 2023-07-03 23:56:51 +08:00
Fangjun Kuang
c3e23ec8d2
Fix logaddexp for ONNX export (#1158) 2023-07-02 10:30:09 +08:00
MicKot
98d89463f6
zipformer2 logaddexp onnx safe (#1157) 2023-06-30 21:16:40 +08:00
Zengwei Yao
ccd8c624dd
support testing onnx exported model on the test sets (#1150)
* support testing onnx exported model on the test sets

* use token_table instead
2023-06-30 12:05:37 +08:00
Desh Raj
c59c89fc13
Minor fix in tedlium results file (#1153) 2023-06-29 13:09:01 +02:00
Wei Kang
db71b03026
Support int8 quantization in decoder (#1152) 2023-06-29 16:48:59 +08:00
Desh Raj
9c2172c1c4
Zipformer for TedLium (#1125)
* initial commit for zipformer tedlium

* fix unk decoding

* add pretrained model and logs

* update for new AsrModel

* add option for choosing rnnt type

* add results with modified rnnt
2023-06-28 16:43:49 +08:00
Fangjun Kuang
968ebd236b
Fix ONNX export of the latest streaming zipformer model. (#1148) 2023-06-27 14:35:59 +08:00
Wei Kang
219bba1310
zipformer wenetspeech (#1130)
* copy files

* update train.py

* small fixes

* Add decode.py

* Fix dataloader in decode.py

* add blank penalty

* Add blank-penalty to other decoding method

* Minor fixes

* add zipformer2 recipe

* Minor fixes

* Remove pruned7

* export and test models

* Replace bpe with tokens in export.py and pretrain.py

* Minor fixes

* Minor fixes

* Minor fixes

* Fix export

* Update results

* Fix zipformer-ctc

* Fix ci

* Fix ci

* Fix CI

* Fix CI

---------

Co-authored-by: Fangjun Kuang <csukuangfj@gmail.com>
2023-06-26 09:33:18 +08:00
frankyoujian
4d5b8369ae
fix small typo (#1144) 2023-06-21 17:17:19 +08:00
Yifan Yang
d667dc365b
Fix for diagnostic (#1135)
* CTC loss return tensor

* Update model.py
2023-06-16 15:04:41 +08:00
Yifan Yang
0a465794a8
Fix Zipformer (#1132)
* Update model.py

* Update train.py

* Update decoder.py
2023-06-15 17:52:14 +08:00
Fangjun Kuang
947f0614c9
Fix running exported model on GPU. (#1131) 2023-06-15 12:25:15 +08:00