Speech recognition using dynamic features of acoustic subword spectra.
Kathy L. BrownV. Ralph AlgaziPublished in: ICASSP (1991)
Keyphrases
- speech recognition
- broadcast news
- speech recognition systems
- acoustic models
- speech sounds
- speech recognizer
- speaker independent
- acoustic features
- speech recognizers
- spoken document retrieval
- speech signal
- automatic speech recognition
- noisy speech
- speech synthesis
- prosodic features
- spectral features
- acoustic signal
- hidden markov models
- spoken language
- n gram
- emotional speech
- vocal tract
- recognition engine
- source localization
- language model
- underwater acoustic
- speech segments
- audio visual
- speaker identification
- pattern recognition
- noisy environments
- hyperspectral
- linear prediction
- visual speech
- speaker recognition
- endpoint detection
- hearing impaired
- formant frequencies