83
Voted
ICASSP
2010
IEEE
15 years 1 months ago
2010
IEEE
We suggest an approach to speech recognition where multiple sides of a conversation in a dialog or meeting are processed and decoded jointly rather than independently. We moreover...