Exploiting
Automatic Speech Recognition (ASR)
knowledge in
Computational Auditory Scene Analysis (CASA)
Dan Ellis
dpwe@icsi.berkeley.edu
International Computer Science Institute, Berkeley CA
Outline:
Analyzing sound mixtures
Integrating ASR into CASA
Inverting speech representation (RASTA)
Preliminary results
Conclusions
Exploiting ASR in CASA -
Dan Ellis
- 1997may21 - 1/9
![[back]](gifs/back.gif)
![[top]](gifs/up.gif)
![[next]](gifs/next.gif)
Analyzing sound mixtures
Integrating ASR into CASA
Inverting speech representation (RASTA)
Preliminary results
Conclusions