CoVoMix2: Advancing Zero-Shot Dialogue Generation with Fully Non-Autoregressive Flow Matching

TL;DR

CoVoMix2 introduces a fully non-autoregressive flow-matching model for zero-shot multi-talker dialogue generation, achieving high speech quality, speaker consistency, and fast inference without relying on intermediate token representations.

Contribution

It presents a novel non-autoregressive framework that directly predicts mel-spectrograms from transcriptions, incorporating speaker disentanglement and masking strategies for improved dialogue synthesis.

Findings

Outperforms MoonCast and Sesame in quality and speed

Operates without transcriptions for prompts

Supports overlapping speech and timing control

Abstract

Generating natural-sounding, multi-speaker dialogue is crucial for applications such as podcast creation, virtual agents, and multimedia content generation. However, existing systems struggle to maintain speaker consistency, model overlapping speech, and synthesize coherent conversations efficiently. In this paper, we introduce CoVoMix2, a fully non-autoregressive framework for zero-shot multi-talker dialogue generation. CoVoMix2 directly predicts mel-spectrograms from multi-stream transcriptions using a flow-matching-based generative model, eliminating the reliance on intermediate token representations. To better capture realistic conversational dynamics, we propose transcription-level speaker disentanglement, sentence-level alignment, and prompt-level random masking strategies. Our approach achieves state-of-the-art performance, outperforming strong baselines like MoonCast and Sesame in speech quality, speaker consistency, and inference speed. Notably, CoVoMix2 operates without requiring transcriptions for the prompt and supports controllable dialogue generation, including overlapping speech and precise timing control, demonstrating strong generalizability to real-world speech generation scenarios.