Multi-resolution cepstral features for phoneme recognition across speech sub-bands.
Paul M. McCourtS. VaseghtNaomi HartePublished in: ICASSP (1998)
Keyphrases
- phoneme recognition
- cepstral features
- multiresolution
- hidden markov models
- speaker dependent
- feature extraction
- automatic speech recognition
- audio signal
- speech recognition
- wavelet transform
- frequency domain
- speaker identification
- discrete wavelet transform
- texture analysis
- subband
- speech signal
- image fusion
- speech corpus
- machine learning
- image processing