Learning latent representations for style control and transfer in end-to-end speech synthesis

Abstract

In this paper, we introduce the Variational Autoencoder (VAE) to an end-to-end speech synthesis model, to learn the latent representation of speaking styles in an unsupervised manner. The style representation learned through VAE shows good properties such as disentangling, scaling, and combination, which makes it easy for style control. Style transfer can be achieved in this framework by first inferring style representation through the recognition network of VAE, then feeding it into TTS network to guide the style in synthesizing speech. To avoid Kullback-Leibler (KL) divergence collapse in training, several techniques are adopted. Finally, the proposed model shows good performance of style control and outperforms Global Style Token (GST) model in ABX preference tests on style transfer.

arXiv commentPaper accepted by ICASSP 2019

References in corpus (6)

• WaveNet: A Generative Model for Raw Audio
• Attention-Based Models for Speech Recognition
• Tutorial on Variational Autoencoders
• Style Tokens: Unsupervised Style Modeling, Control and Transfer in End-to-End Speech Synthesis
• Understanding disentangling in $\beta$-VAE
• Predicting Expressive Speaking Style From Text In End-To-End Speech Synthesis

Cited by in corpus (1)

• Model See Model Do: Speech-Driven Facial Animation with Style Control