Pitch synchronous residual excited speech reconstruction on the MFCC.
Zbynek TychtlJosef PsutkaPublished in: EUSIPCO (2000)
Keyphrases
- speech signal
- speech recognition
- acoustic features
- linear prediction
- formant frequencies
- speaker identification
- fundamental frequency
- automatic speech recognition
- mel frequency cepstral coefficients
- speaker recognition
- speech recognition systems
- image reconstruction
- noisy environments
- speaker verification
- feature extraction
- cepstral coefficients
- non stationary
- audio features
- hidden markov models
- linear predictive
- pattern recognition
- text to speech
- language model
- speech synthesis
- high resolution
- low level
- information retrieval
- reconstruction method
- recognition engine
- vocal tract
- speaker diarization
- speech recognizer
- discrete tomography
- x ray
- visual features
- background noise