Clean speech reconstruction from MFCC vectors and fundamental frequency using an integrated front-end.
Ben MilnerXu ShaoPublished in: Speech Commun. (2006)
Keyphrases
- fundamental frequency
- speech signal
- noisy speech
- speech recognition
- automatic speech recognition
- speaker identification
- speech enhancement
- vocal tract
- linear prediction
- background noise
- hidden markov models
- three dimensional
- back end
- image reconstruction
- noisy environments
- acoustic features
- vector space
- non stationary
- mel frequency cepstral coefficients
- discrete tomography
- speaker recognition
- feature vectors
- cepstral coefficients
- reconstruction method
- high resolution
- multiscale
- information retrieval
- speech synthesis
- face recognition
- information retrieval systems
- probabilistic model
- point distribution
- pattern recognition
- computer vision