Model adaptation based on HMM decomposition for reverberant speech recognition.
Tetsuya TakiguchiSatoshi NakamuraQiang HouKiyohiro ShikanoPublished in: ICASSP (1997)
Keyphrases
- speech recognition
- speech signal
- hidden markov models
- automatic speech recognition
- speaker identification
- language model
- speech recognizer
- noisy environments
- speech processing
- keyword spotting
- speech understanding
- speech recognition systems
- linear prediction
- pattern recognition
- speaker independent
- speech synthesis
- background noise
- handwriting recognition
- speech recognition technology
- vocal tract
- speech retrieval
- speech enhancement
- isolated word recognition
- speech recognition errors
- speaker dependent
- cepstral coefficients
- recognition engine
- probabilistic model
- feature vectors