Refined WaveNet Vocoder for Variational Autoencoder Based Voice Conversion

Abstract

This paper presents a refinement framework of WaveNet vocoders for variational autoencoder (VAE) based voice conversion (VC), which reduces the quality distortion caused by the mismatch between the training data and testing data. Conventional WaveNet vocoders are trained with natural acoustic features but conditioned on the converted features in the conversion stage for VC, and such a mismatch often causes significant quality and similarity degradation. In this work, we take advantage of the particular structure of VAEs to refine WaveNet vocoders with the self-reconstructed features generated by VAE, which are of similar characteristics with the converted features while having the same temporal structure with the target natural features. We analyze these features and show that the self-reconstructed features are similar to the converted features. Objective and subjective experimental results demonstrate the effectiveness of our proposed framework.

arXiv comment5 pages, 7 figures, 1 table. Accepted to EUSIPCO 2019

References in corpus (3)

• Auto-Encoding Variational Bayes
• WaveNet: A Generative Model for Raw Audio
• Voice Conversion Based on Cross-Domain Features Using Variational Auto Encoders

Cited by in corpus (8)

• Voice Conversion for Whispered Speech Synthesis
• Unsupervised Representation Disentanglement using Cross Domain Features and Adversarial Learning in Variational Autoencoder based Voice Conversion
• An Overview of Voice Conversion and its Challenges: From Statistical Modeling to Deep Learning
• A Comparative Study of Self-supervised Speech Representation Based Voice Conversion
• NAUTILUS: a Versatile Voice Cloning System
• Non-parallel Voice Conversion System with WaveNet Vocoder and Collapsed Speech Suppression
• A Cyclical Post-filtering Approach to Mismatch Refinement of Neural Vocoder for Text-to-speech Systems
• Statistical Voice Conversion with Quasi-Periodic WaveNet Vocoder