用非自回归扩散模型加速语音大模型，降低延迟34%以上。
VocalNet-MDM: Accelerating Streaming Speech LLM via Self-Distilled Masked Diffusion Modeling
- 采用分块掩码与自蒸馏技术，解决训练推理不一致和迭代开销问题。
- 仅用6000小时数据训练，推理速度提升3.7到10倍，首块延迟降34%。
- 适合低延迟语音交互场景，兼顾生成质量与实时性，可扩展性强。
近期语音大语言模型在端到端语音交互中表现优异，但主流的自回归范式受限于串行生成，效率低且存在暴露偏差。本文探索非自回归的掩码扩散建模（MDM）用于语音LLM，提出VocalNet-MDM。为适配流式语音交互，解决训练-推理不匹配与迭代开销问题，提出分块层级掩码以对齐训练目标与渐进式掩码状态，并采用迭代自蒸馏将多步精修压缩至更少步骤以实现低延迟推理。在仅6000小时语音数据上训练，VocalNet-MDM相比自回归基线实现3.7×–10×解码加速，首块延迟减少34%，保持良好识别准确率的同时达到顶尖的文本质量和语音自然度，表明MDM是高效、低延迟语音LLM的有前景且可扩展的替代方案。
原文摘要 · Abstract (English)
Recent Speech Large Language Models~(LLMs) have achieved impressive capabilities in end-to-end speech interaction. However, the prevailing autoregressive paradigm imposes strict serial constraints, limiting generation efficiency and introducing exposure bias. In this paper, we investigate Masked Diffusion Modeling~(MDM) as a non-autoregressive paradigm for speech LLMs and introduce VocalNet-MDM. To adapt MDM for streaming speech interaction, we address two critical challenges: training-inference mismatch and iterative overhead. We propose Hierarchical Block-wise Masking to align training objectives with the progressive masked states encountered during block diffusion decoding, and Iterative Self-Distillation to compress multi-step refinement into fewer steps for low-latency inference. Trained on a limited scale of only 6K hours of speech data, VocalNet-MDM achieves a 3.7$\times$--10$\times$ decoding speedup and reduces first-chunk latency by 34\% compared to AR baselines. It maintains competitive recognition accuracy while achieving state-of-the-art text quality and speech naturalness, demonstrating that MDM is a promising and scalable alternative for low-latency, efficient speech LLMs.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。