MFCC computation from magnitude spectrum of higher lag autocorrelation coefficients for robust speech recognition.
Benjamin J. ShannonKuldip K. PaliwalPublished in: INTERSPEECH (2004)
Keyphrases
- speech recognition
- cepstral coefficients
- noisy environments
- hidden markov models
- speech signal
- speech synthesis
- speaker identification
- language model
- speech recognition systems
- pattern recognition
- automatic speech recognition
- spectral subtraction
- speech processing
- speech recognizer
- power spectral
- speech recognition technology
- speaker independent
- speaker dependent
- speech recognition errors
- speech retrieval
- linear prediction
- power spectrum
- digital video library
- non stationary