OneVoice: One Model, Triple Scenarios—Towards Unified Zero-shot Voice Conversion

Project Roadmap

This section will be updated as we progress.

Model Capabilities

OneVoice is a unified model that supports all three voice conversion scenarios: Linguistic-preserving (LVC), Expressive (EVC), and Singing (SVC).

Contents

• Abstract
• Demo: EVC Results
• Demo: SVC Results
• Demo: LVC Results

Abstract

Recent progress of voice conversion~(VC) has achieved a new milestone in speaker cloning and linguistic preservation. But the field remains fragmented, relying on specialized models for linguistic-preserving, expressive, and singing scenarios. We propose OneVoice, a unified zero-shot framework capable of handling all three scenarios within a single model. OneVoice is built upon a continuous language model trained with VAE-free next-patch diffusion, ensuring high fidelity and efficient sequence modeling. Its core design for unification lies in a Mixture-of-Experts (MoE) designed to explicitly model shared conversion knowledge and scenario-specific expressivity. Expert selection is coordinated by a dual-path routing mechanism, including shared expert isolation and scenario-aware domain expert assignment with global-local cues. For precise conditioning, scenario-specific prosodic features are fused into each layer via a gated mechanism, allowing adaptive usage of prosody information. Furthermore, to enable the core idea and alleviate the imbalanced issue (abundant speech vs. scarce singing), we adopt a two-stage progressive training that includes foundational pre-training and scenario enhancement with LoRA-based domain experts. Experiments show that OneVoice matches or surpasses specialized models across all three scenarios, while verifying flexible control over scenarios and offering a fast decoding version as few as 2 steps. Code and model will be released soon.

Demo: Zero-shot EVC Results

Demo: Zero-shot SVC Results

Demo: Zero-shot LVC Results