提前预测对话终点，让语音交互更快更流畅
Endpoint Anticipation for Low-Latency Spoken Dialogue
- 用语音模型提前2.56秒预测对话结束，主动规划后续处理
- 实测平均延迟降低505毫秒，推测性计算仅增28.4%
- 适合需要实时语音交互的复杂系统，如智能助手
低延迟语音对话至关重要，但传统级联架构常受限于反应式回合完成检测。本文提出端点预测（Endpoint Anticipation），将检测方式从被动响应转为主动预判。基于语音的模型可提前最多2.56秒预测对话结束，从而对部分上下文进行大语言模型（LLM）和文本转语音（TTS）管道的推测性执行。我们引入新指标量化延迟降低与计算冗余之间的权衡。在对话和任务导向数据集上的评估显示，该模型持续优于基于语音活动检测（VAP）的基线方法。与Unmute框架集成后，实现平均505毫秒延迟下降，推测性计算增加28.4%，有效掩盖串行瓶颈，支持实时语音到语音交互中的复杂推理。
原文摘要 · Abstract (English)
While low-latency interaction is critical for spoken dialogue, cascaded architectures are often bottlenecked by reactive turn-completion detection. We propose Endpoint Anticipation, shifting from reactive detection to proactive forecasting of end-of-turn signals. Our speech-based model anticipates endpoints upto 2.56 seconds in advance, enabling speculative execution of LLM and TTS pipelines on partial context. We introduce metrics to quantify the trade-off between realized latency reduction and computational redundancy. Evaluation across conversational and task-oriented datasets shows our model consistently outperforms competitive VAP-based baselines. Integration with the Unmute framework demonstrates a 505 ms average latency reduction with a 28.4% increase in speculative computation, effectively masking sequential bottlenecks to enable complex reasoning in real-time speech-to-speech interaction.
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。