DMCA
MAP prediction of pitch from MFCC vectors for speech reconstruction
| Venue: | Proc. ICSLP,2004 |
| Citations: | 1 - 1 self |
BibTeX
@INPROCEEDINGS{Shao_mapprediction,
author = {Xu Shao and Ben Milner},
title = {MAP prediction of pitch from MFCC vectors for speech reconstruction},
booktitle = {Proc. ICSLP,2004},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
This work proposes a method of predicting pitch and voicing from mel-frequency cepstral coefficient (MFCC) vectors. Two maximum a posteriori (MAP) methods are considered. The first models the joint distribution of the MFCC vector and pitch using a Gaussian mixture model (GMM) while the second method also models the temporal correlation of the pitch contour using a combined hidden Markov model (HMM)-GMM framework. Monophone-based HMMs are connected together in the form of an unconstrained monophone grammar which enables pitch to be predicted from unconstrained speech input. Evaluation on 130,000 MFCC vectors reveals a voicing classification accuracy of over 92% and an RMS pitch error of 10Hz. The predicted pitch contour is also applied to MFCC-based speech reconstruction with the resultant speech almost indistinguishable from that reconstructed using a reference pitch. 1.
Keyphrases
mfcc vector map prediction speech reconstruction unconstrained monophone grammar mfcc-based speech reconstruction second method gaussian mixture model classification accuracy rms pitch error predicted pitch contour gmm framework temporal correlation pitch contour joint distribution resultant speech mel-frequency cepstral coefficient first model unconstrained speech input monophone-based hmms combined hidden markov model reference pitch
