4e9bca177a
[ASR] change optimizer and fix import error, test=asr ( #3023 )
...
* mv dataio.py to s2t.io.speechbrain.dataio
mv dataio.py to paddlespeech.s2t.io.speechbrain.dataio
* remove transformers import.
* change optimizer same with released model
* add paddlenlp version in RESULT.md.
* fix run.sh
* fix data.sh step_num.
* add adadelta optimizer config.
* fix wav2vec2 test_wav.sh run error.
* add tokenizer config.
3 years ago
5186319f48
fix load model schedule error, config optional. ( #3008 )
3 years ago
dcf8ef04e0
[ASR] Remove fluid api and useless import, test=asr ( #2944 )
...
* remove fluid api and useless import.
* fix variable name
3 years ago
a8a353d0ac
[ASR] add python simple adadelta optimizer, test=asr ( #2925 )
...
* add simple adeadelta optimizer.
* remove useless log
* remove useless and fluid import.
* add framework.dygraph_only back
3 years ago
004a4d6096
[ASR] rm transformers import and modify variable name consistent with infer.py, test=asr ( #2929 )
...
* rm transformers import and modify variable name consistent with infer.py
* add condition ctc_prefix_beam_search decode.
3 years ago
17a7ebddfa
fix dist_sampler AttributeError ( #2918 )
3 years ago
047092de8e
add wav2vev2_zh aishell recipe, and speechbrain dataloader. ( #2916 )
3 years ago
6728db5b59
[ASR]Whisper remove audio duration limit, test=asr ( #2900 )
3 years ago
64aeb6dccc
remove some fluid api (elementwise_div elementwise_mul sqrt reduce_sum). ( #2859 )
3 years ago
31c2c226ca
clean fluid elementwise_max and square api. ( #2852 )
3 years ago
a99244d86e
fix: whisper language choice, test=asr ( #2828 )
3 years ago
ad40dafa85
fix some bug. ( #2825 )
3 years ago
a8a240d4ef
remove paddle.fluid ( #2740 )
3 years ago
12fa8a2d19
[audio]patch:fix tensor_utils error ( #2738 )
...
* fix tensor utils
3 years ago
42ff946007
[audio] mv paddlespeech/audio to paddleaudio ( #2706 )
...
* split paddlespeech/audio to paddleaudio.
* add sox io ,sox effect, kaldi native fbank to paddleaudio.
3 years ago
c67bf7b4ef
[ASR] support wav2vec2-zh cli, test=asr ( #2697 )
...
* support wav2vec2-zh cli, test=asr
* support wav2vec2-zh cli, test=asr
* support wav2vec2-zh cli, test=asr
* support wav2vec2-zh cli, test=asr
3 years ago
4542684694
[ASR] fix Whisper cli model download path error. test=asr ( #2679 )
...
* add all whisper model size support
* add choices in parser.
* fix Whisper cli model download path error.
* fix resource download path.
* fix code style
3 years ago
fc02cd0540
[doc] update wav2vec2 demos README.md, test=doc ( #2674 )
...
* fix wav2vec2 demos, test=doc
* fix wav2vec2 demos, test=doc
* fix enc_dropout and nor.py, test=asr
3 years ago
b71f1428c7
add all whisper model size support, test=asr ( #2677 )
...
* add all whisper model size support
* add choices in parser.
3 years ago
94a487bd81
[ASR] support wav2vec2 command line and demo ( #2658 )
...
* wav2vec2_cli
* wav2vec2 demo update: support different optimizer and lr_schedular, align mdoel, update input type, test=asr
* wav2vec2 demo update: support different optimizer and lr_schedular, align mdoel, update input type, test=asr
* wav2vec2 demo update: support different optimizer and lr_schedular, align mdoel, update input type, test=asr
* wav2vec2 demo update: support different optimizer and lr_schedular, align mdoel, update input type, test=asr
* Update RESULTS.md
* Update RESULTS.md
* Update base_commands.py
* wav2vec2 demo update: support different optimizer and lr_schedular, align mdoel, update input type, test=asr
* wav2vec2 demo update: support different optimizer and lr_schedular, align mdoel, update input type, test=asr
3 years ago
b1d3f59bcb
[s2t] add whisper asr large model ( #2640 )
...
* add whisper asr large model decoding, test=asr
* fix code style.
* fix json code style.
* remove resource and fix code style.
* fix yapf
* add cli and demos, fix some code style.
* fix some problem by comment.
* fix yapf
3 years ago
8d3494320d
[ASR] wav2vec2_en, test=asr ( #2637 )
...
* wav2vec2_en, test=asr
* wav2vec2_en, test=asr
* wav2vec2_en, test=asr
3 years ago
2c34481ea0
[s2t] quant with wav scp ( #2568 )
...
* add quant hint
* add paddleslim
* using paddleslim 2.3.4 and paddle 2.4
3 years ago
8d3464c050
[s2t] Update wav2vec2 license ( #2600 )
3 years ago
bbf2401e3e
Merge pull request #2524 from zh794390558/u2
...
[speechx] add u2/u2pp asr inference
3 years ago
eac545e1db
Merge pull request #2544 from Zth9730/fix_attention
...
[s2t] fix attention eval bug, do not compose kv in infer
3 years ago
1ea828c30e
fix attention val bug
3 years ago
964c22c677
Merge pull request #2532 from Zth9730/wav2vec2.0
...
[s2t] fix wav2vec2 report loss bug
3 years ago
86f65f0b8e
fix wav2vec2 report loss bug
3 years ago
f1ca564731
Merge pull request #2518 from Zth9730/wav2vec2.0
...
[ASR] wav2vec2 ASR, pre-trained wav2vec2 based CTC for librispeech
3 years ago
2ae94bd277
freeze wav2vec2=True, change loss report and update README.md
3 years ago
3d994f5c23
format wav2vec2 demo
3 years ago
19180d359d
format wav2vec2 demo
3 years ago
6e429f0513
support wav2vec2ASR on librispeech
3 years ago
290c23b9d7
add u2 nnet, u2 nnet main, codelab, and can compile
3 years ago
cda440e6f0
use reverse_weight in decode.yaml
3 years ago
c98b5dd173
fix masked_fill which will nan in trainning
3 years ago
9277fcb8a8
fix attn can not train
3 years ago
1f4f98b171
fix bug
3 years ago
e86337a423
fix bug
3 years ago
925abcca23
format
3 years ago
2a75405e9a
Merge branch 'develop' into u2pp_export
3 years ago
3ed24474d2
wenetspeech asr1 quant
3 years ago
467cfd4e75
Merge pull request #2489 from Zth9730/u2++_server
...
[ASR] support u2pp based cli and server, optimiz code of u2pp decode (reversed_weight parameter)
3 years ago
5bbe6e9897
support u2pp cli and server, optimiz code of u2pp decode, test=asr
3 years ago
bdf876ea7b
Merge branch 'develop' into u2pp_export
3 years ago
afda7ed7d1
remove useless code
3 years ago
b20bf7d5de
masked_fill by multiply, remove while
3 years ago
feb27e2a84
fuse linear kv
3 years ago
3adb20b468
eliminate shape and slice
3 years ago
46088c0a16
elimiate attn transpose
3 years ago
f9e3eaa024
transpose in matmul
3 years ago
3d7ca93861
bool type slice
3 years ago
c2c8a662b1
refactor reshape
3 years ago
6de81d74d9
elimiete cast dtype for bool op
3 years ago
8e7a315e00
remove comment
3 years ago
b7388ce25a
eliminate useless unsqueese
3 years ago
1a1ce92cb4
Merge pull request #2415 from Zth9730/u2++_decoder
...
[s2t] support bitransformer decoder
3 years ago
d3e5937591
support bitransformer decoder
3 years ago
7382050e21
fix bug on win
3 years ago
d25871a7b0
format
3 years ago
b10512eb0e
more config or u2pp
3 years ago
00b2c1c8fb
fix forward attention decoder caller
3 years ago
309c8d70d9
add reverse weight
3 years ago
9b66680ea4
Merge branch 'u2++_decoder' into u2pp_export
3 years ago
027535dec1
support bitransformer decoder, test=asr
3 years ago
0a95689461
support bitransformer decoder
3 years ago
455379b88e
support bitransformer decoder
3 years ago
1a56a6e42b
add bitransformer decoder, test=asr
3 years ago
53d6baff0b
format
3 years ago
549d477592
fix code style
3 years ago
4d5cfd4003
export param from cnofig
3 years ago
e3298c79ce
Merge branch 'develop' into u2_export
3 years ago
260752aa2a
using forward_attention_decoder
3 years ago
8690a00bd8
add feature pipeline layer(cmvn, fbank), but to_static and jit.layer output is not equal
3 years ago
3a8869fba4
rm to_static decarator; configure jit save for ctc_activation
3 years ago
1c9f238ba0
configurable export
3 years ago
63aeb747b0
more comment
3 years ago
d638325c46
do not jit save forward; using slice for zeros([0,0,0,0]) tensor
3 years ago
663e3ab58e
fix dp init
3 years ago
6745e9dd6b
fix dp init
3 years ago
598eb1a5ef
Merge branch 'develop' into fix_dp_init
3 years ago
9560d650db
fix dp init
3 years ago
82e04d7815
fix trianer
3 years ago
2bb40c41ba
Merge pull request #2351 from Zth9730/fix_deepspeech
...
[s2t] fix deepspeech2 decode_wav
3 years ago
ab92e2c98c
fix deepspeech2 decode_wav
3 years ago
795eb7bd10
format paddlespeech with pre-commit ( #2331 )
3 years ago
cdcb1a5316
s2t: fix encoder.py
3 years ago
ed2819d7af
fix format test=asr
3 years ago
ed80b0e2c3
fix multigpu training test=asr
3 years ago
733ec7f2bc
fix conformer multi-gpu training test=asr
3 years ago
c1fbfe928e
add test
3 years ago
05bc258833
update docstring
3 years ago
6149daa221
export ctc_activation
3 years ago
060e337623
fix dataloader factory, test=asr
3 years ago
812d80ab1c
Merge branch 'develop' of https://github.com/PaddlePaddle/DeepSpeech into u2_export
3 years ago
e5a6c243f1
fix jit save for conformer
3 years ago
4e7106d9e2
Support dy2st
3 years ago
ef37f73a01
fix cnn cache dy2st shape
3 years ago
e21cceea51
Remove blank line
3 years ago
zxcd
cxumol
YangZhou
Zth9730
Hui Zhang
tianhao zhang
TianYuan
huangyuxin
0x45f