GANSpeech: Adversarial Training for High-Fidelity Multi-Speaker Speech Synthesis

by   Jinhyeok Yang, et al.

Recent advances in neural multi-speaker text-to-speech (TTS) models have enabled the generation of reasonably good speech quality with a single model and made it possible to synthesize the speech of a speaker with limited training data. Fine-tuning to the target speaker data with the multi-speaker model can achieve better quality, however, there still exists a gap compared to the real speech sample and the model depends on the speaker. In this work, we propose GANSpeech, which is a high-fidelity multi-speaker TTS model that adopts the adversarial training method to a non-autoregressive multi-speaker TTS model. In addition, we propose simple but efficient automatic scaling methods for feature matching loss used in adversarial training. In the subjective listening tests, GANSpeech significantly outperformed the baseline multi-speaker FastSpeech and FastSpeech2 models, and showed a better MOS score than the speaker-specific fine-tuned FastSpeech2.



There are no comments yet.


page 3


Speaker-adaptive neural vocoders for statistical parametric speech synthesis systems

This paper proposes speaker-adaptive neural vocoders for statistical par...

ECAPA-TDNN for Multi-speaker Text-to-speech Synthesis

In recent years, neural network based methods for multi-speaker text-to-...

Multi-Speaker End-to-End Speech Synthesis

In this work, we extend ClariNet (Ping et al., 2019), a fully end-to-end...

A Text-to-Speech Pipeline, Evaluation Methodology, and Initial Fine-Tuning Results for Child Speech Synthesis

Speech synthesis has come a long way as current text-to-speech (TTS) mod...

Generative Adversarial Network based Speaker Adaptation for High Fidelity WaveNet Vocoder

Neural networks based vocoders, typically the WaveNet, have achieved spe...

Building Synthetic Speaker Profiles in Text-to-Speech Systems

The diversity of speaker profiles in multi-speaker TTS systems is a cruc...

Combining Adversarial Training and Disentangled Speech Representation for Robust Zero-Resource Subword Modeling

This study addresses the problem of unsupervised subword unit discovery ...
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.