研究人员开发了全双工口语模型（FD-SLMs）的新框架和基准，该模型可以同时进行听和说。DuplexAct引入了一个包含三级动作分类和真实对话数据的基准，用于评估FD-SLMs中的交互动作。D2框架将现有的半双工模型转换为低延迟的高性能FD-SLMs，在VoiceBench等基准上取得了最先进的性能。此外，JAL-Turn通过联合建模声学和语言线索，为对话系统中的实时轮次检测提供了一种轻量级的纯语音方法。 AI
影响 全双工口语模型的这些进步可能带来更自然、响应更快的语音AI代理和虚拟助手。
排序理由 该集群包含三篇arXiv论文，介绍了口语模型的新基准和框架。
AI 生成摘要 · Google Gemini · 来自 3 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
arXiv:2610.17738v1 Announce Type: new Abstract: End-to-end (E2E) full-duplex spoken language models (FD-SLMs) enable simutanous listening and speaking continously, requiring models to take appropriate interaction actions as conversations unfold. However, existing benchmarks lack …
arXiv:2610.17124v1 Announce Type: cross Abstract: Full-duplex spoken language models (FD-SLMs) support simultaneous listening and speaking, but often lag behind half-duplex spoken language models (HD-SLMs) in general capabilities. We propose Dialogue-to-Duplex (D2), a framework t…
arXiv:2603.26515v2 Announce Type: replace-cross Abstract: Efficient and robust turn-taking detection remains a significant challenge in industrial-grade voice AI agents. Existing systems often rely on acoustic or linguistic cues alone, limiting accuracy and stability, whereas dir…