A Survey on Neural Speech Synthesis

TL;DR

This survey reviews recent advances in neural speech synthesis, covering key components, challenges, and future trends to guide researchers and industry practitioners in developing more natural and efficient TTS systems.

Contribution

It provides a comprehensive overview of neural TTS, including key components, advanced topics, resources, and future research directions, which was lacking in existing literature.

Findings

Summarizes recent neural TTS techniques and their improvements.

Highlights challenges like low-resource and robust TTS.

Provides resources and future research directions.

Abstract

Text to speech (TTS), or speech synthesis, which aims to synthesize intelligible and natural speech given text, is a hot research topic in speech, language, and machine learning communities and has broad applications in the industry. As the development of deep learning and artificial intelligence, neural network-based TTS has significantly improved the quality of synthesized speech in recent years. In this paper, we conduct a comprehensive survey on neural TTS, aiming to provide a good understanding of current research and future trends. We focus on the key components in neural TTS, including text analysis, acoustic models and vocoders, and several advanced topics, including fast TTS, low-resource TTS, robust TTS, expressive TTS, and adaptive TTS, etc. We further summarize resources related to TTS (e.g., datasets, opensource implementations) and discuss future research directions. This survey can serve both academic researchers and industry practitioners working on TTS.