On desensitizing the Mel-Cepstrum to spurious spectral components for Robust Speech Recognition.
Vivek TyagiChristian WellekensPublished in: ICASSP (1) (2005)
Keyphrases
- speech recognition
- cepstral coefficients
- linear predictive coding
- noisy environments
- speech signal
- hidden markov models
- speech processing
- pattern recognition
- language model
- automatic speech recognition
- speech recognizer
- speech understanding
- speech synthesis
- speaker independent
- speech recognition technology
- speech recognition systems
- feature set
- speaker identification
- speaker adaptation
- speech recognition errors
- speaker recognition
- speaker dependent
- isolated word
- character recognition
- keyword spotting