[论文] Training Intelligent Voice Assistant Wakeup with Controllable Syntheti...

论文概要

研究领域: ML
作者: Marcin Sowański, Kacper Leszczyński, Kacper Krzywicki, Krzysztof Wodnicki
发布时间: 2026-09-25
arXiv: 2609.27037

中文摘要

唤醒词检测是虚拟助手的关键组件，是无缝用户交互的入口。本文提出一种新型唤醒系统，将传统直接关键词检测扩展为上下文触发检测：初始唤醒词激活后，系统通过推理区分用户指令与无关语音，确保高效且具备上下文感知能力的响应。我们提出一种数据生成架构，可产生 62.3 小时的可控多说话人对话语料，包含直接调用、上下文接续与未指向助手的语音。实验结果证明了该方法在多种合成交谈场景中的有效性。我们开源代码、数据集与训练模型，以促进可复现性与智能助手技术的进一步发展。

原文摘要

Wake word detection is a critical component of virtual assistants, serving as the gateway to seamless user interactions. This paper introduces a novel wake-up system that extends traditional direct keyword detection with contextual trigger detection. After an initial wake word activation, the system uses reasoning to distinguish between user commands and unrelated speech, ensuring efficient and context-aware engagement. We present a data generation architecture that produces a 62.3-hour corpus of controllable multi-speaker conversations containing direct invocations, contextual follow-ups, and non-addressed speech. Experimental results demonstrate the effectiveness of the proposed approach across diverse synthetic conversational scenarios. We release the code, dataset and trained models to...

自动采集于 2026-09-25

#论文 #arXiv #ML #小凯