Hui Zhang
dd06472432
extract ctc_align into ctc_utils
3 years ago
Hui Zhang
8b45c3e65e
refactor trainer.py and rm ueseless dir setup code
3 years ago
Hui Zhang
eb4b389262
more log; refactor ctc decoders; rm useless code
3 years ago
Hui Zhang
ce3af793c1
Merge pull request #872 from Jackwaterveg/Hub
...
change test chain and g2p
3 years ago
Hui Zhang
2691013232
ctcloss can work w/ paddle2.1.2, but loss larger than before
3 years ago
Hui Zhang
86a221d58f
fix autolog add cuda device env when not has
3 years ago
Hui Zhang
b291c69386
add checkpoint to save parameters
3 years ago
Hui Zhang
37563d975e
ds2 model_type more info
3 years ago
Hui Zhang
8977f1850e
close editdistance package format warning
3 years ago
Hui Zhang
466672e1de
no_sync if paddle support else nullcontext
3 years ago
Hui Zhang
b4e16eb815
revert ctc_utils that can work with paddle 2.1.2 and new
3 years ago
Hui Zhang
913ea33f29
remove conv and rnn uesless
3 years ago
Hui Zhang
f4f2d6f07e
print deps module version
3 years ago
Hui Zhang
c8e96d732b
bool logical, sum and multiply op; ctc grad norm; support old and new pd api
3 years ago
Hui Zhang
d05baeb6b0
update ted zh en
3 years ago
Jackwaterveg
dfa0db5cc3
Merge pull request #870 from PaddlePaddle/timit
...
fix timit scripts; reader filtype case;
3 years ago
Hui Zhang
c5daa34e7c
Merge pull request #869 from Jackwaterveg/Hub
...
add input checker for hub
3 years ago
Hui Zhang
251d32a609
fix timit scripts; reader filtype case;
3 years ago
huangyuxin
ebde17af20
add input checker for hub
3 years ago
Junkun
46df01151f
Merge branch 'develop' of https://github.com/LittleChenCc/DeepSpeech into develop
3 years ago
Hui Zhang
e0f91b0994
Merge pull request #865 from PaddlePaddle/collator
...
fix collator test
3 years ago
Hui Zhang
856d641c9c
multi worker for dataloader
3 years ago
Hui Zhang
b7b1bda34f
test refactor collator
3 years ago
Jackwaterveg
00d7654273
Merge pull request #864 from PaddlePaddle/collator
...
refactor st and asr collator
3 years ago
Hui Zhang
32aaf403b5
Merge pull request #861 from PaddlePaddle/ctc
...
more ctc check; valid dataloader with num workers
3 years ago
Hui Zhang
f628e21816
refactor kaldi/tarfile loader, st and asr collator
3 years ago
Junkun
c32cb734a6
update the result of TED-EN-ZH
3 years ago
Hui Zhang
77c38dd521
Merge pull request #862 from Jackwaterveg/Hub
...
Add bin for hub(test single wav file)
3 years ago
huangyuxin
1a46125175
add bin for hub
3 years ago
Hui Zhang
f7d7e70cb2
more ctc check; valid dataloader with num workers
3 years ago
huangyuxin
9c37d10992
optimize the log
3 years ago
Jackwaterveg
4b225b7602
Merge pull request #858 from PaddlePaddle/ctc
...
ctc using nn.Dropout; ds2 libri vald batch_size / 4
3 years ago
Hui Zhang
0c08915207
ctc using nn.Dropout; ds2 libri vald batch_size / 4
3 years ago
huangyuxin
30b3e237e2
optimize the 1xt2x
3 years ago
huangyuxin
6dc32d185b
Merge branch 'develop' of https://github.com/PaddlePaddle/DeepSpeech into release_model
3 years ago
huangyuxin
08281eca72
fix bug: bug of space id in score.cpp, add detokenize
3 years ago
huangyuxin
db831ae702
fix the text_featurize
3 years ago
huangyuxin
cb49aba64f
fix the ctc.cpp bug
3 years ago
huangyuxin
f6c9976189
Merge branch 'develop' of https://github.com/PaddlePaddle/DeepSpeech into release_model
3 years ago
Hui Zhang
30563981f0
replace space when build vocab
3 years ago
Hui Zhang
86f34784e3
fix import
3 years ago
Hui Zhang
4c098a0f33
explit space as <space>
3 years ago
huangyuxin
4a1d3ad9b3
fix the bug of parallel using condition
3 years ago
huangyuxin
285e0c9cad
merge the change
3 years ago
Hui Zhang
8e16315ada
librispeech s1 support multi process decode and sclite
3 years ago
Hui Zhang
83a5e0f8d6
using jsonlines to read manifest and dump decode result
3 years ago
Hui Zhang
33b8790252
refactor avg_model; fix set_value not support start==end
3 years ago
Hui Zhang
9d5eb74066
fix decode json file
3 years ago
Hui Zhang
ae87bc8c7a
dump decode result as jsonlines
3 years ago
Hui Zhang
c6e8a33b73
fix set_device; more utils; args.opts support multi same name
3 years ago
huangyuxin
264bba760b
fix the bug: read space as unk
3 years ago
Hui Zhang
913b2300c3
nprocs 0 for cpu, other for gpu
3 years ago
Hui Zhang
98b15eda05
batch WaveDataset
3 years ago
Hui Zhang
ecb5d4f862
Merge pull request #847 from PaddlePaddle/resume_train
...
resume train with epoch and iteration increase
3 years ago
Hui Zhang
bab29b94f1
fix train log
3 years ago
huangyuxin
7e96942c58
Merge branch 'develop' of https://github.com/PaddlePaddle/DeepSpeech into release_model
3 years ago
huangyuxin
f0184352f5
change the code format to 2.x style
3 years ago
Hui Zhang
56e55c2171
not save ckpt when except, since resume train will increase epoch and step
3 years ago
Hui Zhang
431106b986
fix bugs
3 years ago
Hui Zhang
676d160dfa
more resume ckpt info
3 years ago
Hui Zhang
daaa72a606
resuem train with epoch and iteration increase
3 years ago
Hui Zhang
7775abd727
fix prof switch
3 years ago
Hui Zhang
6ae6ca85b7
more prof log
3 years ago
Hui Zhang
7ae204eb0f
fix log for u2
3 years ago
Hui Zhang
282914f45b
Merge branch 'develop' into fix_varbase
3 years ago
Hui Zhang
231499a347
VarBase.__getitem__ work for np.int64, np.longlong; but __setitem_varbase__ not support paddle.int16/set_value op not support
3 years ago
Hui Zhang
000183ea49
tensor.size to tensor.shape
3 years ago
Hui Zhang
cd001d5daf
change type_as to astype
3 years ago
Hui Zhang
16e60160f8
Kaldi ( #839 )
...
* can do frames, real stft
* format
* stft complex, powspec, magspec
* add common utils
* add window process func
* using frames and matmul as stft
* read with 2d; window process
* test with dither, remove dc offset, preermphs
* add doc string
* more frontend utils
* add logspec
* fix typing
* add delpoy mergify label
3 years ago
Hui Zhang
9fb349f935
fix benchmark cli
3 years ago
Hui Zhang
054e099b28
format
3 years ago
Hui Zhang
0e91d26ae3
fix log; add report to trainer
3 years ago
Hui Zhang
6de20de3f8
rename reporter.scope to ObsScope
3 years ago
Hui Zhang
576e94da04
log interval 1 when benchmark
3 years ago
Hui Zhang
d7a33b9d5e
update config with contextlib
3 years ago
Hui Zhang
cda6ca8323
add benchmark flags, and logic
3 years ago
Hui Zhang
c75082295a
varbase getitem support np.longlong since paddle 2.2.0RC
3 years ago
Hui Zhang
a997b5a61c
rename ckpt suffix to np
3 years ago
Hui Zhang
16b9c33deb
format
3 years ago
Hui Zhang
3a5258f6a0
lr and opt param will restore from ckpt, so we do not set lr manully
3 years ago
Hui Zhang
7907319288
fix profiler
3 years ago
Hui Zhang
5fdda953b9
add op profiling
3 years ago
Hui Zhang
256e9c1b9c
more doc for egs
3 years ago
Hui Zhang
3843372958
u2 with chianer updater
3 years ago
Hui Zhang
2480be8ebc
timer info for st,u2 kaldi
3 years ago
Hui Zhang
85674d22d4
fix bugs
3 years ago
Hui Zhang
f0470e0529
not dump all grad info, since slow down train porcess
3 years ago
Hui Zhang
8fef667f80
fix u2 ctc config bugs
3 years ago
Hui Zhang
8873ebe38c
add timer for u2; refactor grad norm type
3 years ago
Hui Zhang
890a28f9bf
add more ctc conf
3 years ago
Hui Zhang
65e666378d
add timer info
3 years ago
Jackwaterveg
59766313b1
Merge pull request #817 from PaddlePaddle/ctc
...
export ctc grad norm config
3 years ago
Jackwaterveg
7d0204e9fe
Merge pull request #816 from PaddlePaddle/bool_mul
...
float_mul_bool type promote, rhs type promote to lhs type
3 years ago
Hui Zhang
1a8c5278a1
export ctc grad norm config
3 years ago
Hui Zhang
6e4a3aff07
float_mul_bool type promote, lhs type promote to rhs type, https://github.com/PaddlePaddle/Paddle/pull/29265
3 years ago
Hui Zhang
c29ee83a46
add timer
3 years ago
Hui Zhang
244132c1c4
fix activation
3 years ago
Hui Zhang
7e136d0893
support no_sync for backward; ds support accum grad
3 years ago
Hui Zhang
184d30dd9c
relase librispeech audio max len to 30 second
3 years ago
huangyuxin
04d9db199f
add blank_id parameter
3 years ago