Voice conversion using partial least squares regression

Authors:
Elina Helander;Tuomas Virtanen;Jani Nurminen;Moncef Gabbouj
Affiliations:
Department of Signal Processing, Tampere University of Technology, Tampere, Finland;Department of Signal Processing, Tampere University of Technology, Tampere, Finland;Nokia Devices R&D, Tampere, Finland;Department of Signal Processing, Tampere University of Technology, Tampere, Finland
Venue:
IEEE Transactions on Audio, Speech, and Language Processing
Year:
2010

Citing 8
Cited 1

Neural networks and the bias/variance dilemma

Neural Computation
Restructuring speech representations using a pitch-adaptive time-frequency smoothing and an instantaneous-frequency-based F0 extraction: possible role of a repetitive structure in sounds

Speech Communication
Scalable Techniques from Nonparametric Statistics for Real Time Robot Learning

Applied Intelligence
Conversion Function Clustering and Selection Using Linguistic and Spectral Information for Emotional Voice Conversion

IEEE Transactions on Computers
Analysis of Speaker Adaptation Algorithms for HMM-Based Speech Synthesis and a Constrained SMAPLR Adaptation Algorithm

IEEE Transactions on Audio, Speech, and Language Processing
Nonparallel training for voice conversion based on a parameter adaptation approach

IEEE Transactions on Audio, Speech, and Language Processing
A Spectral Conversion Approach to Single-Channel Speech Enhancement

IEEE Transactions on Audio, Speech, and Language Processing
Voice Conversion Based on Maximum-Likelihood Estimation of Spectral Parameter Trajectory

IEEE Transactions on Audio, Speech, and Language Processing

Voice conversion using linear prediction coefficients and artificial neural network

Proceedings of the CUBE International Information Technology Conference

Quantified Score

Hi-index	0.00

Visualization

Abstract

Voice conversion can be formulated as finding a mapping function which transforms the features of the source speaker to those of the target speaker. Gaussian mixture model (GMM)- based conversion is commonly used, but it is subject to overfitting. In this paper, we propose to use partial least squares (PLS)-based transforms in voice conversion. To prevent overfitting, the degrees of freedom in the mapping can be controlled by choosing a suitable number of components. We propose a technique to combine PLS with GMMs, enabling the use of multiple local linear mappings. To further improve the perceptual quality of the mapping where rapid transitions between GMM components produce audible artefacts, we propose to low-pass filter the component posterior probabilities. The conducted experiments show that the proposed technique results in better subjective and objective quality than the baseline joint density GMM approach. In speech quality conversion preference tests, the proposed method achieved 67% preference score against the smoothed joint density GMM method and 84% preference score against the unsmoothed joint density GMM method. In objective tests the proposed method produced a lower Melcepstral distortion than the reference methods.