动态多专家投影器与稳定路由用于多语言语音识别
Dynamic Multi-Expert Projectors with Stabilized Routing for Multilingual Speech Recognition
AI总结：
SMEAR-MoE通过稳定多专家投影器实现多语言语音识别的高效跨语言共享与性能提升。
摘要：
近年来，基于大语言模型（LLM）的自动语音识别（ASR）通过轻量级投影器将冻结的语音编码器与LLMs连接起来。尽管在单语设置中有效，但单一投影器难以捕捉多语言ASR所需的多样化的语音到语义映射。为此，我们提出了SMEAR-MoE，一种稳定混合专家投影器，确保梯度流向所有专家，防止专家崩溃，同时实现跨语言共享。我们系统地比较了单体、静态多投影器和动态MoE设计在四种印地语系语言（印地语、马拉地语、泰米尔语、泰卢固语）上的表现。我们的SMEAR-MoE实现了优异的性能，相比单一投影器基线，实现了高达7.6%的相对WER下降，同时保持了可比的运行时效率。专家路由的分析进一步显示了语言上有意义的专业化，相关语言共享专家。这些结果表明，稳定的多专家投影器是实现可扩展且稳健的多语言ASR的关键。
英文摘要：
Recent advances in LLM-based ASR connect frozen speech encoders with Large Language Models (LLMs) via lightweight projectors. While effective in monolingual settings, a single projector struggles to capture the diverse acoustic-to-semantic mappings required for multilingual ASR. To address this, we propose SMEAR-MoE, a stabilized Mixture-of-Experts projector that ensures dense gradient flow to all experts, preventing expert collapse while enabling cross-lingual sharing. We systematically compare monolithic, static multi-projector, and dynamic MoE designs across four Indic languages (Hindi, Marathi, Tamil, Telugu). Our SMEAR-MoE achieves strong performance, delivering upto a 7.6% relative WER reduction over the single-projector baseline, while maintaining comparable runtime efficiency. Analysis of expert routing further shows linguistically meaningful specialization, with related languages sharing experts. These results demonstrate that stable multi-expert projectors are key to scalable and robust multilingual ASR.
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。
