端到端追踪说话人，让长对话转写带时间戳和身份标签。
G-STAR: End-to-End Global Speaker-Tracking Attributed Recognition
- 用缓存驱动的追踪模块与语音大模型联动，同步处理时间边界与说话人身份。
- 在分段推理下，跨片段说话人一致性准确率显著提升。
- 适合需要高精度多说话人对话转写的场景，如会议记录、访谈分析。
我们研究长时序、多说话人重叠场景下的带时间戳说话人标注自动语音识别（SA-ASR）。在此设定中，分块推理需保持会话级说话人身份一致，并生成带时间戳、说话人标签的转录文本。现有语音大模型系统往往侧重局部聚类或全局标记，难以联合建模精细的时间边界与跨块身份关联。本文提出 G-STAR，一个端到端框架，将缓存条件化的说话人追踪模块与语音大模型转录主干耦合。追踪模块提供带时间定位的结构化说话人线索，语音大模型基于这些线索生成带属性的文本。G-STAR 支持组件独立优化与联合端到端训练，可在异构监督与领域漂移下灵活学习。在分块解码协议下，无论是基于理想分段的局部评估还是完整会议的全局评估，均表现出强的说话人标注转录性能。
原文摘要 · Abstract (English)
We study timestamped speaker-attributed automatic speech recognition (SA-ASR) for long-form, multi-party speech with overlap. In this setting, chunk-wise inference must preserve meeting-level speaker identity consistency while producing time-stamped, speaker-labeled transcripts. Prior Speech-LLM systems tend to prioritize either local diarization or global labeling, lacking the ability to jointly model fine-grained temporal boundaries and robust cross-chunk identity linking. We propose G-STAR, an end-to-end framework that couples a cache-conditioned speaker-tracking module with a Speech-LLM transcription backbone. The tracker provides structured speaker cues with temporal grounding, and the LLM generates attributed text conditioned on these cues. G-STAR supports component-wise optimization and joint end-to-end training, enabling flexible learning under heterogeneous supervision and domain shift. Under chunk-wise decoding protocols, experiments on both oracle-segmented local evaluation and full-meeting global evaluation show strong speaker-attributed transcription performance.
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。