基于动态语言分组的MoE：通过层次化路由增强语码转换语音识别
Dynamic Language Group-Based MoE: Enhancing Code-Switching Speech Recognition with Hierarchical Routing
AI总结：
本文提出DLG-MoE，一种基于动态语言分组的MoE模型，通过层次化路由机制显式建模语言属性并隐式协调专家路由，有效解决了CS-ASR任务并充分利用了参数扩展优势，在性能和灵活性上超越现有方法。
摘要：
混合专家模型是处理语码转换语音识别（CS-ASR）任务的一种有前景的方法。然而，现有的基于MoE的CS-ASR工作尚未充分利用MoE参数扩展能力的优势。本文提出DLG-MoE，一种基于动态语言分组的MoE，能够有效处理CS-ASR任务并利用参数扩展的优势。DLG-MoE基于层次化路由机制运行。首先，语言路由器显式建模语言属性，并将表示分发到相应的语言专家组。随后，每个语言组内的无监督路由器隐式建模语言之外的属性，协调专家路由与协作。DLG-MoE在CS-ASR任务上优于现有MoE方法，同时展现出极大的灵活性。它支持不同的top-$k$推理和流式处理能力，还能灵活剪枝模型参数以获得单语子模型。代码已发布。
英文摘要：
The Mixture of Experts (MoE) model is a promising approach for handling code-switching speech recognition (CS-ASR) tasks. However, the existing CS-ASR work on MoE has yet to leverage the advantages of MoE's parameter scaling ability fully. This work proposes DLG-MoE, a Dynamic Language Group-based MoE, which can effectively handle the CS-ASR task and leverage the advantages of parameter scaling. DLG-MoE operates based on a hierarchical routing mechanism. First, the language router explicitly models the language attribute and dispatches the representations to the corresponding language expert groups. Subsequently, the unsupervised router within each language group implicitly models attributes beyond language and coordinates expert routing and collaboration. DLG-MoE outperforms the existing MoE methods on CS-ASR tasks while demonstrating great flexibility. It supports different top-$k$ inference and streaming capabilities and can also prune the model parameters flexibly to obtain a monolingual sub-model. The code has been released.
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。
