“Zhiyuan Tang 2016-04-18”版本间的差异
来自cslt Wiki
| 第5行： | 第5行： | ||
1. enhancing the joint model with SWBD focusing on speech recognition, shows improvement[http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=tangzy&step=view_request&cvssid=515]; | 1. enhancing the joint model with SWBD focusing on speech recognition, shows improvement[http://192.168.0.51:5555/cgi-bin/cvss/cvss_request.pl?account=tangzy&step=view_request&cvssid=515]; | ||
| − | 2. problem remains that when the WSJ was reduced to be 8k, the advantage of joint training disappeared at least for speech recognition | + | 2. problem remains that when the WSJ was reduced to be 8k, the advantage of joint training disappeared at least for speech recognition. (comment: WSJ was reduced to 8k by mistake, so the pipeline needs to be reconducted). |
| − | + | ||
2016年4月18日 (一) 08:51的版本
Last week:
1. enhancing the joint model with SWBD focusing on speech recognition, shows improvement[1];
2. problem remains that when the WSJ was reduced to be 8k, the advantage of joint training disappeared at least for speech recognition. (comment: WSJ was reduced to 8k by mistake, so the pipeline needs to be reconducted).
This week:
1. find the reason why joint training failed on 8k WSJ;
2. more experiemnts for refining the joint model, such as enhancing the enhanced model again with speaker data;
2. following ICASSP 16.
