“Sinovoice-2014-01-13”版本间的差异
来自cslt Wiki
（→6000 hour 16k trainin） |
|||
| 第12行： | 第12行： | ||
==Corpora== | ==Corpora== | ||
| − | * | + | * 60 hour data were cutted |
| + | * Just sending out for labeling | ||
| + | * Waiting for out-source platform construction | ||
| 第34行： | 第36行： | ||
* Lexicon and LM were done | * Lexicon and LM were done | ||
* Making MFCC features | * Making MFCC features | ||
| + | * Initial model can be delivered before the spring holiday | ||
=DNN Decoder= | =DNN Decoder= | ||
* Initial trail of DNN decoder based on the Sinovoice code was failed, largely due to FST compiler | * Initial trail of DNN decoder based on the Sinovoice code was failed, largely due to FST compiler | ||
* Change the strategy to an integrated approach: use the sinovoice system to control connections, and use Kaldi base for asr engine | * Change the strategy to an integrated approach: use the sinovoice system to control connections, and use Kaldi base for asr engine | ||
2014年1月13日 (一) 07:01的版本
目录
Project management
- Xiaoming and Xiao Na were added into the mail list
- Potential Huawei conference-transcribing project was discussed
DNN training
Environment setting
- New disk space (3T) was created and mounted at /nfs/disk1
- Jobs with 100 threads work fine on the cluster
Corpora
- 60 hour data were cutted
- Just sending out for labeling
- Waiting for out-source platform construction
470 hour 8k training
- CE training done
- MPE training partially done
| Model | CE | MPE1 | MPE2 | MPE3 | MPE4 |
|---|---|---|---|---|---|
| 4k states | 23.27/22.85 | 21.35/18.87 | 21.18/18.76 | 21.07/18.54 | |
| 8k states | 22.16/22.22 | - | 20.36/17.94 | - |
6000 hour 16k training
- Audio files done. File with incorrect sampling rates were removed
- Lexicon and LM were done
- Making MFCC features
- Initial model can be delivered before the spring holiday
DNN Decoder
- Initial trail of DNN decoder based on the Sinovoice code was failed, largely due to FST compiler
- Change the strategy to an integrated approach: use the sinovoice system to control connections, and use Kaldi base for asr engine
