Msdtron: a high-capability multi-speaker speech synthesis system for diverse data using characteristic information

TL;DR

This paper introduces Msdtron, a multi-speaker speech synthesis system that leverages characteristic speech information and novel neural components to better model diverse speaker data, significantly improving synthesis quality.

Contribution

The paper presents Msdtron, a novel multi-speaker speech synthesis system with a new excitation spectrogram representation and conditional gated LSTM for enhanced modeling of speaker diversity.

Findings

Reduced mel-spectrogram reconstruction error

Significant improvement in subjective speaker adaptation quality

Effective handling of diverse speaker data

Abstract

In multi-speaker speech synthesis, data from a number of speakers usually tend to have great diversity due to the fact that the speakers may differ largely in ages, speaking styles, emotions, and so on. It is important but challenging to improve the modeling capabilities for multi-speaker speech synthesis. To address the issue, this paper proposes a high-capability speech synthesis system, called Msdtron, in which 1) a representation of the harmonic structure of speech, called excitation spectrogram, is designed to directly guide the learning of harmonics in mel-spectrogram. 2) conditional gated LSTM (CGLSTM) is proposed to control the flow of text content information through the network by re-weighting the gates of LSTM using speaker information. The experiments show a significant reduction in reconstruction error of mel-spectrogram in the training of the multi-speaker model, and a great improvement is observed in the subjective evaluation of speaker adapted model.