msb-public
/
PaddleSpeech
mirror of https://github.com/PaddlePaddle/PaddleSpeech
1
0
Code Issues Projects Releases Wiki Activity
You can not select more than 25 topics Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
Branches Tags
${ item.name }
Create tag ${ searchTerm }
Create branch ${ searchTerm }
from '62c21e951f'
${ noResults }
PaddleSpeech/examples/aishell/asr3/RESULT.md

19 lines
541 B
Raw Normal View History Unescape

[ASR] add wav2vec2 aishell model result, test=asr (#3012) * Create RESULT.md * add wav2vec2ASR-large-aishell1 finetune model. * update model link and add readme. * fix released model info.
2 years ago
# AISHELL
## Version
* paddle version: develop (commit id: daea892c67e85da91906864de40ce9f6f1b893ae)
* paddlespeech version: develop (commit id: c14b4238b256693281e59605abff7c9435b3e2b2)
[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.
2 years ago
* paddlenlp version: 2.5.2
[ASR] add wav2vec2 aishell model result, test=asr (#3012) * Create RESULT.md * add wav2vec2ASR-large-aishell1 finetune model. * update model link and add readme. * fix released model info.
2 years ago
## Device
* python: 3.7
* cuda: 10.2
* cudnn: 7.6
## Result
train: Epoch 80, 2*V100-32G, batchsize:5
| Model | Params | Config | Augmentation| Test set | Decode method | WER |
| --- | --- | --- | --- | --- | --- | --- |
| wav2vec2ASR | 324.49 M | conf/wav2vec2ASR.yaml | spec_aug | test-set | greedy search | 5.1009 |