智能语音设备协同唤醒响应机制研究与实现

doi:10.19784/j.cnki.issn1672-0172.2026.02.011

家电科技 ›› 2026, Vol. 0 ›› Issue (2): 71-75.doi: 10.19784/j.cnki.issn1672-0172.2026.02.011

智能语音设备协同唤醒响应机制研究与实现

刘永红¹, 徐翼¹, 伍云云¹, 张文彬¹, 吴洋¹, 吴则昊^1,2

1.美的集团（上海）有限公司上海 201700;
2.东北大学佛山研究生创新学院广东佛山 528000

出版日期:2026-04-01 发布日期:2026-06-17
通讯作者: 徐翼,E-mail：xuyi42@midea.com。
作者简介:刘永红,硕士学位。研究方向：边缘智能。地址：上海市青浦区西虹桥商务区美的全球创新园区。E-mail：yonghong2.liu@midea.com。

Research and implementation of collaborative wake-up response mechanism for voice-enabled intelligent devices

Liu Yonghong¹, Xu Yi¹, Wu Yunyun¹, Zhang Wenbin¹, Wu Yang¹, Wu Zehao^1,2

1. Midea Group (Shanghai) Co., Ltd. Shanghai 201700;
2. Foshan Graduate School of Innovation Northeastern University Foshan 528000

Online:2026-04-01 Published:2026-06-17

摘要/Abstract

摘要： 随着智能语音设备在家居环境中的广泛部署,多设备因同时响应用户唤醒指令而引发的交互冲突问题日益突出。由于现有唤醒协同响应机制依赖网络基础设施,在离线场景下失效,且通过信号强度等指标难以准确识别用户直面的目标设备。所以提出了一种基于IEEE 802.11管理帧的无网络协同方案,通过设备间直接通信,融合相干混响信噪比与双阶段信噪比等多维唤醒特征,实现分布式决策。实验结果表明,方案在模拟家居环境下实现96.65%的设备唯一准确率和90.18%的直面准确率,显著提升了多设备语音交互的准确性与用户体验。

关键词: 语音交互, 协同响应, 分布式决策, 设备直接通信, 相干混响信噪比

Abstract: With the widespread deployment of voice-enabled intelligent devices in home environments, interaction conflicts caused by multiple devices simultaneously responding to user's wake-up commands have become increasingly prominent. This is because the existing wake-up collaborative response mechanisms rely on network infrastructure and fail in offline scenarios; moreover, it is difficult to accurately identify the user's “facing” target device using metrics such as signal strength alone. To address this, a network-free cooperative scheme is proposed. Based on IEEE 802.11 management frames, the scheme enables direct device-to-device communication and fuses multidimensional acoustic features—including coherent reverberant SNR and two-stage SNR—to support distributed decision-making. Experimental results in simulated home environments show a 96.65% unique-response rate and 90.18% facing-device accuracy, significantly improving the accuracy and user experience of multi-device voice interactions.

Key words: Voice interaction, Collaborative mechanisms, Distributed decision-making, Direct communication, Coherent reverberant signal-to-noise ratio

中图分类号:

TM925
TP391

刘永红, 徐翼, 伍云云, 张文彬, 吴洋, 吴则昊. 智能语音设备协同唤醒响应机制研究与实现[J]. 家电科技, 2026, 0(2): 71-75.

Liu Yonghong, Xu Yi, Wu Yunyun, Zhang Wenbin, Wu Yang, Wu Zehao. Research and implementation of collaborative wake-up response mechanism for voice-enabled intelligent devices[J]. Journal of Appliance Science & Technology, 2026, 0(2): 71-75.

/ 推荐

参考文献

[1] Mahamuni C.Exploring IoT-Applications: A Survey of Recent Progress, Challenges, and Impact of AI, Blockchain, and Disruptive Technologies[C]. 7th International Conference on Electronics, Communication and Aerospace Technology (ICECA 2023), 2023.
[2] 冯维思. 基于深度学习的智能家居现状分析[C]// 中国通信工业协会教育专委会暨中国通信工业协会, 信息安全与云计算校企联盟2023年年会论文集, 2023.
[3] Schönherr L, Golla M, Eisenhofer T, et al.Exploring accidental triggers of smart speakers[J]. Computer speech & language, 2022(73): 101328.
[4] Meng-Schneider N, Kostas R, Vaniea K, et al. Multi-User Smart Speakers - A Narrative Review of Concerns and Problematic Interactions[C]. Extended Abstracts of the2023 CHI Conference on Human Factors in Computing Systems (CHI EA '23), 2023.
[5] 陶梦春, 毛跃辉, 汪进, 等. 就近语音唤醒及识别关键技术研究与空调应用[J]. 家电科技, 2021(4): 110-113.
[6] Ciccarelli G, Barber J, Nair A, et al.Challenges and Opportunities in Multi-device Speech Processing[C]. The 23rd Annual Conference of the International Speech Communication Association (INTERSPEECH 2022), 2022.
[7] Barber J, Fan Y, Zhang T.End-To-End Alexa Device Arbitration[C]. IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2022), 2022.
[8] Benítez-Guijarro A, Callejas Z, Noguera M, et al.Coordination of Speech Recognition Devices in Intelligent Environments with Multiple Responsive Devices[C]. 13th International Conference on Ubiquitous Computing and Ambient Intelligence (UCAmI 2019), 2019.
[9] 沈靖雯, 俞贵涛, 刘俊翔, 等. 多设备云语音优先级控制系统设计[J]. 家电科技, 2021(增刊): 146-150.
[10] Liu S, Feng L, Gong Y, et al.Deep learning based stage-wise two-dimensional speaker localization with large ad-hoc microphone arrays[EB/OL]. https://arxiv.org/abs/2210.10265.
[11] IEEE Standard for Information technology--Telecommunications and information exchange between systems Local and metropolitan area networks--Specific requirements - Part 11: Wireless LAN Medium Access Control (MAC) and Physical Layer (PHY) Specifications[S]. IEEE STD 802.11-2020, 2021.
[12] Schwarz A,Kellermann W.Coherent-to-Diffuse Power Ratio Estimation for Dereverberation[J]. IEEE/ACM Transactions on Audio, Speech and Language Processing, 2015(23):1006-1018.
[13] Brendel A, Kellermann W.Distributed Source Localization in Acoustic Sensor Networks Using the Coherent-to-Diffuse Power Ratio[J]. IEEE Journal of Selected Topics in Signal Processing, 2019(13): 61-75.
[14] Xiang Q, Lei T, Pan C, et al.On Multichannel Coherent-to-Diffuse Power Ratio Estimation[J]. IEEE Sensors Journal, 2024(24): 37455-37462.

智能语音设备协同唤醒响应机制研究与实现

Research and implementation of collaborative wake-up response mechanism for voice-enabled intelligent devices

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 9

Metrics

本文评价

推荐阅读 10

[1]	潘悦然, 焦利敏, 杨荣祥, 崔庆用, 姚昌松, 徐翼. 智慧家庭语音交互意图理解评测现状[J]. 家电科技, 2025, 0(zk): 17-23.
[2]	唐杰, 贾巨涛, 李立辉, 张鹏, 杨驰. 智能语音交互中的多意图解析技术研究[J]. 家电科技, 2024, 0(zk): 418-421.
[3]	陈伟杰, 吴振樑, 胡志娟, 熊军, 韩东, 陈妃味, 仵小勇, 陈春好. 离线语音自由说关键技术研究及空调应用[J]. 家电科技, 2024, 0(zk): 425-429.
[4]	焦利敏, 刘泽超, 顾子谦, 金轮, 胡亚欣, 王生泽, 刘冬阳. 智能家电语音交互能力测试语料库建设的研究[J]. 家电科技, 2022, 0(zk): 130-134.
[5]	贾巨涛, 张鹏, 唐杰, 吴伟, 詹培旋. 智能语音交互中的语义引导回复技术研究[J]. 家电科技, 2022, 0(zk): 608-611.
[6]	沈靖雯, 俞贵涛, 刘俊翔, 秦臻, 沈素丽. 多设备云语音优先级控制系统设计[J]. 家电科技, 2021, 0(zk): 146-150.
[7]	李红伟, 焦利敏, 郑杰昌, 冯长卿, 胡亚欣, 赵燕伟. 智能家电远程操控及语音交互安全性研究[J]. 家电科技, 2020, 0(zk): 27-31.
[8]	李明杰, 贾巨涛, 宋德超, 吴伟, 韩林峄. 一种基于少量训练数据的口语语义理解技术[J]. 家电科技, 2020, 0(zk): 222-224.
[9]	吕秀凤, 焦其意, 方波. 综述人工智能在冰箱上的应用[J]. 家电科技, 2018, 0(1): 36-37.