Non-linear spectral subtraction (NSS) and hidden Markov models for robust speech recognition in car noise environments.
Philip LockwoodJérôme BoudyMarc BlanchetPublished in: ICASSP (1992)
Keyphrases
- speech recognition
- spectral subtraction
- hidden markov models
- speech signal
- noisy environments
- noisy speech
- automatic speech recognition
- speech recognizer
- speech processing
- speech synthesis
- speaker identification
- keyword spotting
- speaker independent
- language model
- speech enhancement
- speech recognition systems
- visual speech recognition
- neural network
- discriminative training
- conditional random fields
- pattern recognition