通过说话者无关活动流扩展多说话者ASR
Scaling Multi-Talker ASR with Speaker-Agnostic Activity Streams
AI总结：
针对多说话者ASR推理成本随说话者数量上升的问题，提出将说话者活动转换为两条说话者无关流的方法，结合新启发式规则，在保持性能的同时大幅缩短运行时间。
摘要：
多说话者自动语音识别（ASR）日益普遍的训练范式是利用说话者活动信号调整单说话者ASR模型以处理重叠语音。这类系统虽有效，但需为每个说话者运行一次ASR模型，导致推理成本随说话者数量增加而上升，限制了实用性。本研究提出一种方法，通过将特定说话者的活动输出转换为两条说话者无关流，使活动条件ASR系统的推理成本与说话者数量解耦。核心挑战在于，直接将说话者活动合并为流会严重降低识别效果，因为预训练ASR模型假设输入是连续的单说话者语音。为解决此问题，我们设计了新的启发式规则，旨在保持对话连续性并维持与现有系统的兼容性。实验表明，该方法与Diarization-Conditioned Whisper（DiCoW，说话者分割条件化Whisper）兼容，可在AMI和ICSI会议数据集上大幅缩短运行时间，同时保持有竞争力的性能。
英文摘要：
An increasingly common training paradigm for multi-talker automatic speech recognition (ASR) is to use speaker activity signals to adapt single-speaker ASR models for overlapping speech. Although effective, these systems require running the ASR model once per speaker, resulting in inference costs that scale with the number of speakers and limiting their practicality. In this work, we propose a method that decouples the inference cost of activity-conditioned ASR systems from the number of speakers by converting speaker-specific activity outputs into two speaker-agnostic streams. A central challenge is that naïvely merging speaker activities into streams significantly degrades recognition, since pretrained ASR models assume contiguous, single-speaker inputs. To address this, we design new heuristics aimed at preserving conversational continuity and maintaining compatibility with existing systems. We show that our approach is compatible with Diarization-Conditioned Whisper (DiCoW) to greatly reduce runtimes on the AMI and ICSI meeting datasets while retaining competitive performance.
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。
