CSSinger: End-to-End Chunkwise Streaming Singing Voice Synthesis System Based on Conditional Variational Autoencoder

TL;DR

This paper introduces CSSinger, an end-to-end chunkwise streaming singing voice synthesis system based on conditional variational autoencoders, achieving high-quality, expressive singing voice generation with low latency.

Contribution

It presents the first fully end-to-end streaming SVS system using latent VAE representations, improving performance and addressing latency issues in practical applications.

Findings

Achieves high expressiveness and pitch accuracy in synthesized singing voices.

Demonstrates effective streaming inference with low latency.

Outperforms traditional systems in quality and responsiveness.

Abstract

Singing Voice Synthesis (SVS) aims to generate singing voices of high fidelity and expressiveness. Conventional SVS systems usually utilize an acoustic model to transform a music score into acoustic features, followed by a vocoder to reconstruct the singing voice. It was recently shown that end-to-end modeling is effective in the fields of SVS and Text to Speech (TTS). In this work, we thus present a fully end-to-end SVS method together with a chunkwise streaming inference to address the latency issue for practical usages. Note that this is the first attempt to fully implement end-to-end streaming audio synthesis using latent representations in VAE. We have made specific improvements to enhance the performance of streaming SVS using latent representations. Experimental results demonstrate that the proposed method achieves synthesized audio with high expressiveness and pitch accuracy in both streaming SVS and TTS tasks.