TTS-by-TTS: TTS-driven Data Augmentation for Fast and High-Quality Speech Synthesis

TL;DR

This paper introduces a TTS-driven data augmentation approach that uses an autoregressive TTS system to generate large-scale synthetic data, significantly enhancing the quality of non-autoregressive speech synthesis models.

Contribution

The paper presents a novel data augmentation method leveraging AR TTS to improve non-AR TTS quality, especially with limited training data.

Findings

Synthetic data increased from 5 to 179 hours.

Achieved 40% higher mean opinion score compared to conventional methods.

Significant quality improvement demonstrated through perceptual tests.

Abstract

In this paper, we propose a text-to-speech (TTS)-driven data augmentation method for improving the quality of a non-autoregressive (AR) TTS system. Recently proposed non-AR models, such as FastSpeech 2, have successfully achieved fast speech synthesis system. However, their quality is not satisfactory, especially when the amount of training data is insufficient. To address this problem, we propose an effective data augmentation method using a well-designed AR TTS system. In this method, large-scale synthetic corpora including text-waveform pairs with phoneme duration are generated by the AR TTS system and then used to train the target non-AR model. Perceptual listening test results showed that the proposed method significantly improved the quality of the non-AR TTS system. In particular, we augmented five hours of a training database to 179 hours of a synthetic one. Using these databases, our TTS system consisting of a FastSpeech 2 acoustic model with a Parallel WaveGAN vocoder achieved a mean opinion score of 3.74, which is 40% higher than that achieved by the conventional method.