Robust speech recognition via modeling spectral coefficients with HMM's with complex Gaussian components.
William J. J. RobertsSadaoki FuruiPublished in: INTERSPEECH (2000)
Keyphrases
- speech recognition
- cepstral coefficients
- hidden markov models
- speech signal
- noisy environments
- speaker adaptation
- automatic speech recognition
- linear prediction
- keyword spotting
- language model
- speech understanding
- speech processing
- speech recognizer
- speech synthesis
- pattern recognition
- speech recognition technology
- speaker independent
- speaker identification
- speech recognition systems
- spectral subtraction
- speech recognizers
- neural network
- handwriting recognition
- speaker dependent
- speech recognition errors
- speech retrieval
- subband
- natural language