Reducing one-to-many problem in Voice Conversion by equalizing the formant locations using dynamic frequency warping

TL;DR

This paper proposes a method using dynamic frequency warping to equalize formant locations in voice conversion, effectively reducing the one-to-many problem and improving speech quality.

Contribution

It introduces a novel formant equalization technique with dynamic frequency warping to address the one-to-many problem in voice conversion.

Findings

Significant improvement in speech quality observed

Formant equalization reduces over-smoothing effects

Method effectively addresses one-to-many problem

Abstract

In this study, we investigate a solution to reduce the effect of one-to-many problem in voice conversion. One-to-many problem in VC happens when two very similar speech segments in source speaker have corresponding speech segments in target speaker that are not similar to each other. As a result, the mapper function usually over-smoothes the generated features in order to be similar to both target speech segments. In this study, we propose to equalize the formant location of source-target frame pairs using dynamic frequency warping in order to reduce the complexity. After the conversion, another dynamic frequency warping is further applied to reverse the effect of formant location equalization during the training. The subjective experiments showed that the proposed approach improves the speech quality significantly.