利用语音推测解码加速自回归语音合成推理
Accelerating Autoregressive Speech Synthesis Inference With Speech Speculative Decoding
AI总结：
本文提出语音推测解码（SSD）框架，通过轻量级草稿模型生成候选token序列并由目标模型并行验证，在保持高保真度和自然度的同时实现1.4倍的自回归语音合成推理加速。
摘要：
现代利用语言模型的自回归语音合成模型已展现出卓越性能。然而，这些模型中下一token预测的序列化特性导致了显著延迟，阻碍了它们在推理速度至关重要的场景中的部署。在本工作中，我们提出了语音推测解码（Speech Speculative Decoding，SSD），这是一种用于加速自回归语音合成的新型框架。具体而言，我们的方法采用一个轻量级草稿模型来生成候选token序列，随后由目标模型使用所提出的SSD框架并行验证这些序列。实验结果表明，与传统自回归解码相比，SSD实现了1.4倍的显著加速，同时保持了高保真度和自然度。主观评估进一步验证了SSD在加速推理的同时保持目标模型感知质量的有效性。
英文摘要：
Modern autoregressive speech synthesis models leveraging language models have demonstrated remarkable performance. However, the sequential nature of next token prediction in these models leads to significant latency, hindering their deployment in scenarios where inference speed is critical. In this work, we propose Speech Speculative Decoding (SSD), a novel framework for autoregressive speech synthesis acceleration. Specifically, our method employs a lightweight draft model to generate candidate token sequences, which are subsequently verified in parallel by the target model using the proposed SSD framework. Experimental results demonstrate that SSD achieves a significant speedup of 1.4x compared with conventional autoregressive decoding, while maintaining high fidelity and naturalness. Subjective evaluations further validate the effectiveness of SSD in preserving the perceptual quality of the target model while accelerating inference.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
