用扩散语言模型提升语音识别准确率，支持并行生成与联合解码。
Diffusion Language Models for Speech Recognition
- 引入掩码与均匀状态扩散模型用于语音识别重评分。
- 联合解码结合CTC与扩散模型，显著提升识别准确率。
- 代码开源，适合语音识别与扩散模型研究者使用。
扩散语言模型因其双向注意力和并行文本生成能力，成为主流语言模型的有力替代。本文探索其在语音识别中的应用，提出将掩码扩散语言模型（MDLM）与均匀状态扩散模型（USDM）用于重评分的完整方案。设计一种新联合解码方法：在每一步解码中，融合CTC输出的帧级概率分布与USDM计算的词级概率分布，生成结合强语言知识与声学信息的新候选。实验表明，USDM与MDLM均能显著提升识别准确率。所有代码与训练方案均已公开。
原文摘要 · Abstract (English)
Diffusion language models have recently emerged as a leading alternative to standard language models, due to their ability for bidirectional attention and parallel text generation. In this work, we explore variants for their use in speech recognition. Specifically, we introduce a comprehensive guide to incorporating masked diffusion language models (MDLM) and uniform-state diffusion models (USDMs) for rescoring ASR hypotheses. Additionally, we design a new joint-decoding method that combines CTC and USDM by integrating the framewise probability distributions derived from CTC with the labelwise probability distributions computed by USDM at each decoding step, thereby generating new candidates that combine strong language knowledge from USDM and acoustic information from CTC. Our findings reveal that USDM, as well as MDLM, can significantly improve the accuracy of recognized text. We publish all our code and recipes.
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。