“2013-04-12”版本间的差异
来自cslt Wiki
| 第1行： | 第1行： | ||
1. Data sharing | 1. Data sharing | ||
| − | + | (1) Acoustic data ready. PLP feature + HLDA (with LDA transform). Model is PLP+HLDA+ML+MPE. All software ready. Good. | |
| + | |||
| + | (2)The LM data and model are being transferred. | ||
2. DNN progress | 2. DNN progress | ||
| − | Chao | + | (1) 400hour Tencent data training ready. MFCC+LDA (300/1200/1200/1220/40/1200/38xx), followed by (MFCC+BN) X LDA. |
| + | (2) comparision between MFCC and BN with fmpe. Relative improvement is xx%. | ||
| + | (3) BN system and hybrind system: relative perforamnce. | ||
| + | (4) GPU and CPU style comparision: still on progress. Data checking. SGE is still problematic (Chao can help). | ||
| + | (5) RTF comparision between DNN hybrid and GMM: | ||
| + | |||
3.Kaldi/HTK merge | 3.Kaldi/HTK merge | ||
| − | + | (1) HTK2Kaldi: problematic. HMM structure is erroratic. Need to modify the toolkit in Kaldi. | |
| − | + | (2) Kaldi2HTK: need to design a new tool. | |
4. Embedded progress | 4. Embedded progress | ||
| 第17行： | 第24行： | ||
(2). Prototype design. Application design is on going. Plan to deliver DNN decoder. Sparse DNN might be a good solution. | (2). Prototype design. Application design is on going. Plan to deliver DNN decoder. Sparse DNN might be a good solution. | ||
| + | |||
| + | (3). QA LM training, word files ready. trying to start the LM training. Refer to the doc Chao provided. | ||
| + | /nfs/asrhome/asr/lm/chs.lm/lm.qa | ||
2013年4月12日 (五) 06:07的版本
1. Data sharing
(1) Acoustic data ready. PLP feature + HLDA (with LDA transform). Model is PLP+HLDA+ML+MPE. All software ready. Good.
(2)The LM data and model are being transferred.
2. DNN progress
(1) 400hour Tencent data training ready. MFCC+LDA (300/1200/1200/1220/40/1200/38xx), followed by (MFCC+BN) X LDA. (2) comparision between MFCC and BN with fmpe. Relative improvement is xx%. (3) BN system and hybrind system: relative perforamnce. (4) GPU and CPU style comparision: still on progress. Data checking. SGE is still problematic (Chao can help). (5) RTF comparision between DNN hybrid and GMM:
3.Kaldi/HTK merge
(1) HTK2Kaldi: problematic. HMM structure is erroratic. Need to modify the toolkit in Kaldi. (2) Kaldi2HTK: need to design a new tool.
4. Embedded progress
(1). GFCC testing. noisy robust while not very good in silence.
(2). Prototype design. Application design is on going. Plan to deliver DNN decoder. Sparse DNN might be a good solution.
(3). QA LM training, word files ready. trying to start the LM training. Refer to the doc Chao provided. /nfs/asrhome/asr/lm/chs.lm/lm.qa
