arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.04016cs.CLcs.AI
收藏

面向实时语音智能体的泰语语义话轮结束检测

Thai Semantic End-of-Turn Detection for Real-Time Voice Agents

SCBX(SCBX集团)

Thanapol Popit, Natthapath Rungseesiripak, Monthol Charattrakool, Saksorn Ruangtanusak

AI总结:

针对传统音频话轮结束检测延迟高、鲁棒性差的问题,本文首次系统性研究纯文本泰语话轮结束检测,对比多种模型方法,建立泰语基准,证明小型微调模型可实现低延迟检测,适用于端侧实时语音智能体。

摘要:

流畅的语音交互需要可靠且低延迟地检测用户何时结束讲话。传统的基于音频静音的端点检测会增加数百毫秒的延迟,且在用户犹豫或存在语言特有现象时会失效。据我们所知,本文首次针对实时智能体开展了纯文本泰语话轮结束(EOT)检测的系统性研究。我们对比了紧凑型大语言模型(LLM)的零样本和少样本提示方法,与轻量Transformer的监督微调方法。利用YODAS语料库的转录字幕及泰语特有语言线索(如句末助词),我们将EOT检测建模为词元边界上的二分类决策。我们报告了清晰的准确率-延迟权衡,并提供了可公开使用的实现方案。本工作建立了泰语EOT检测基准,证明小型微调模型可实现近乎即时的EOT决策,适用于端侧智能体。

英文摘要:

Fluid voice-to-voice interaction requires reliable and low-latency detection of when a user has finished speaking. Traditional audio-silence end-pointers add hundreds of milliseconds of delay and fail under hesitations or language-specific phenomena. We present, to our knowledge, the first systematic study of Thai text-only end-of-turn (EOT) detection for real-time agents. We compare zero-shot and few-shot prompting of compact LLMs to supervised fine-tuning of lightweight transformers. Using transcribed subtitles from the YODAS corpus and Thai-specific linguistic cues (e.g., sentence-final particles), we formulate EOT as a binary decision over token boundaries. We report a clear accuracy-latency tradeoff and provide a public-ready implementation plan. This work establishes a Thai baseline and demonstrates that small, fine-tuned models can deliver near-instant EOT decisions suitable for on-device agents.

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。

↑