Interfacing Sound Stream Segregation to Automatic Speech Recognition - Preliminary Results on Listening to Several Sounds Simultaneously.
Hiroshi G. OkunoTomohiro NakataniTakeshi KawabataPublished in: AAAI/IAAI, Vol. 2 (1996)
Keyphrases
- automatic speech recognition
- acoustic features
- speech signal
- environmental sounds
- sound source
- speech recognition
- audio stream
- word error rate
- hidden markov models
- sound signals
- broadcast news
- conversational speech
- speech retrieval
- data streams
- speaker verification
- music information retrieval
- spontaneous speech
- spoken words
- vocal tract
- word recognition
- musical instrument
- speech synthesis
- recognition errors
- noisy environments
- visual features
- computer vision
- neural network
- non stationary
- speech corpus
- information retrieval systems