SoulX-Singer: Towards High-Quality Zero-Shot Singing Voice Synthesis

TL;DR

SoulX-Singer is a high-quality, open-source singing voice synthesis system that excels in zero-shot generalization, supports multiple languages and musical controls, and is accompanied by a dedicated benchmark for evaluation.

Contribution

The paper introduces SoulX-Singer, a novel SVS system with practical deployment features, multilingual support, and a new benchmark for zero-shot evaluation.

Findings

Achieves state-of-the-art synthesis quality across languages.

Supports controllable singing generation with symbolic and melodic inputs.

Provides a systematic zero-shot evaluation benchmark.

Abstract

While recent years have witnessed rapid progress in speech synthesis, open-source singing voice synthesis (SVS) systems still face significant barriers to industrial deployment, particularly in terms of robustness and zero-shot generalization. In this report, we introduce SoulX-Singer, a high-quality open-source SVS system designed with practical deployment considerations in mind. SoulX-Singer supports controllable singing generation conditioned on either symbolic musical scores (MIDI) or melodic representations, enabling flexible and expressive control in real-world production workflows. Trained on more than 42,000 hours of vocal data, the system supports Mandarin Chinese, English, and Cantonese and consistently achieves state-of-the-art synthesis quality across languages under diverse musical conditions. Furthermore, to enable reliable evaluation of zero-shot SVS performance in practical scenarios, we construct SoulX-Singer-Eval, a dedicated benchmark with strict training-test disentanglement, facilitating systematic assessment in zero-shot settings.