“2014年12月28日-SGR”版本间的差异
来自cslt Wiki
| （相同用户的一个中间修订版本未显示） | |||
| 第6行： | 第6行： | ||
6. [important] spatial histogram equalization. MAS-HEQ<br/> | 6. [important] spatial histogram equalization. MAS-HEQ<br/> | ||
7. blind spatial subtraction.<br/> | 7. blind spatial subtraction.<br/> | ||
| − | 8. denoising DAE<br/> | + | 8. denoising DAE. multi-speaker.<br/> |
| + | 9. trainseg scale conjugate gradient.<br/> | ||
| + | 10. [to text group] RNN RAE.<br/> | ||
| + | 11. whisper<br/> | ||
| + | 12. [1272 to text group]mixture-model vMF<br/> | ||
| + | 13. [1241 to wang]<br/> | ||
| + | 14. [1240 to xiangyu]<br/> | ||
| + | 15. [1161 to xiangyu]<br/> | ||
| + | 16. [1148 to wang]<br/> | ||
| + | 17. [1057] ultranc? boundary<br/> | ||
| + | 18. [1038 to wang & yinshi]<br/> | ||
| + | 19. [1032 to xuewei] wiener filtering<br/> | ||
| + | 20. [1026 to xuewei]<br/> | ||
2014年12月28日 (日) 08:45的最后版本
1. speech transformed to image, classification
2. [pass] MFCC+
3. EGG usage. insertion is not important
4. co-training. GMM-HMM adaption. class-based LM adaptation.
5. [conflict] CNN.tone information. F0, FFV.
6. [important] spatial histogram equalization. MAS-HEQ
7. blind spatial subtraction.
8. denoising DAE. multi-speaker.
9. trainseg scale conjugate gradient.
10. [to text group] RNN RAE.
11. whisper
12. [1272 to text group]mixture-model vMF
13. [1241 to wang]
14. [1240 to xiangyu]
15. [1161 to xiangyu]
16. [1148 to wang]
17. [1057] ultranc? boundary
18. [1038 to wang & yinshi]
19. [1032 to xuewei] wiener filtering
20. [1026 to xuewei]
