arXiv：cs.AI· Lichen Zhu, Yueqian Lin, Yiheng Wang, Hai "Helen" Li, Yiran Chen·· 6 小时前AI 评分47
PersonaPlex-7B 全双工语音模型对话转向时机研究
Coupled but Late: Turn-Taking Between Full-Duplex Speech Models in Unscripted Dialogue
AI 导读
两个 PersonaPlex-7B 实例在共享时钟上交换音频 token 进行无脚本对话，其转向时机呈现耦合，但轮换偏晚，中位数为 400-560 ms，而人类为 137 ms。在对伙伴发言最后 120 ms 的处理上，人类投射约十分之一的转向，而模型仅 1%。延迟单向信道会使响应一比一平移，且响应前的准备期为空，符合感知到结束后的反应式等待，而非人类时机所需的转向结束投射。
正文
Abstract:Full-duplex speech models are trained to converse with a person, but they are increasingly made to converse with each other, in self-play data generation, agent societies, and model-based evaluation. In that loop no human absorbs a timing error: each model's turn-taking is the other's input. We ask what timing the loop settles into. Two PersonaPlex-7B instances exchange audio tokens on a shared clock in unscripted conversation, and one floor-transfer rule is applied to them and to Switchboard. Their timing is coupled: re-pairing speakers across conversations destroys it. But the floor changes hands late, at a median of 400-560 ms against 137 ms for humans, and the last 120 ms of the partner's turn, where human projection places a tenth of its transfers, holds 1% of theirs. Delaying one direction of the channel shifts the response one-for-one and leaves the run-up to it empty, consistent with a reactive wait after the perceived end rather than the turn-end projection human timing requires.
| Comments: | The paper is under review |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.08683 [cs.AI] |
| (or arXiv:2610.08683v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.08683 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yiheng Wang [view email]
[v1]
Tue, 6 Oct 2026 17:03:54 UTC (69 KB)
来源：arXiv：cs.AI · arxiv.org