An Empirical Study on End-to-End Singing Voice Synthesis with Encoder-Decoder Architectures

TL;DR

This paper investigates the use of encoder-decoder neural network architectures combined with vocoders to improve the quality and efficiency of end-to-end singing voice synthesis, enabling non-experts to generate natural singing voices.

Contribution

It introduces an end-to-end neural singing voice synthesis framework using encoder-decoder models trained with pitch, lyrics, and beat data, demonstrating high-quality results.

Findings

Models produce smooth, clear, and natural singing voices.

Training with pitch, lyrics, and beat data enhances synthesis quality.

End-to-end approach simplifies singing voice creation for non-experts.

Abstract

With the rapid development of neural network architectures and speech processing models, singing voice synthesis with neural networks is becoming the cutting-edge technique of digital music production. In this work, in order to explore how to improve the quality and efficiency of singing voice synthesis, in this work, we use encoder-decoder neural models and a number of vocoders to achieve singing voice synthesis. We conduct experiments to demonstrate that the models can be trained using voice data with pitch information, lyrics and beat information, and the trained models can produce smooth, clear and natural singing voice that is close to real human voice. As the models work in the end-to-end manner, they allow users who are not domain experts to directly produce singing voice by arranging pitches, lyrics and beats.