Speech recognition in nonstationary noise based on parallel HMMs and spectral subtraction.
Ryuji MineTetsunori KobayashiKatsuhiko ShiraiPublished in: Systems and Computers in Japan (1996)
Keyphrases
- spectral subtraction
- speech recognition
- non stationary
- hidden markov models
- speech signal
- automatic speech recognition
- noisy environments
- pattern recognition
- language model
- speech synthesis
- speech recognizer
- noisy speech
- autoregressive
- speech recognition systems
- random fields
- speaker identification
- background noise
- vocal tract
- speaker independent
- linear prediction
- acoustic models
- information retrieval