Paper Info
Title
Modulation Spectrum-Constrained Trajectory Training Algorithm For Gmm-Based Voice Conversion
Abstract
This paper presents a novel training algorithm for Gaussian Mixture Model (GMM) -based Voice Conversion (VC). One of the advantages of GMM-based VC is computationally efficient conversion processing enabling to achieve real-time VC applications. On the other hand, the quality of the converted speech is still significantly worse than that of natural speech. In order to address this problem while preserving the computationally efficient conversion processing, the proposed training method enables 1) to use a consistent optimization criterion between training and conversion and 2) to compensate a Modulation Spectrum (MS) of the converted parameter trajectory as a feature sensitively correlated with over-smoothing effects causing quality degradation of the converted speech. The experimental results demonstrate that the proposed algorithm yields significant improvements in term of both the converted speech quality and the conversion accuracy for speaker individuality compared to the basic training algorithm.
Year
DOI
Venue
2015
10.1109/ICASSP.2015.7178894
2015 IEEE INTERNATIONAL CONFERENCE ON ACOUSTICS, SPEECH, AND SIGNAL PROCESSING (ICASSP)
Keywords
Field
DocType
GMM-based voice conversion, over-smoothing, modulation spectrum. training algorithm
Hafnium,Pragmatics,Pattern recognition,Computer science,Speech quality,Algorithm,Speech recognition,Artificial intelligence,Mixture model,Trajectory,Modulation spectrum
Conference
ISSN
Citations 
PageRank 
1520-6149
7
0.44
References 
Authors
19
4
Name
Order
Citations
PageRank
Shinnosuke Takamichi17522.08
Tomoki Toda21874167.18
Alan W. Black34391742.28
Satoshi Nakamura41099194.59