郑旭琪
（广东电网有限责任公司揭阳供电局，广东 揭阳 522000）
摘要：随着电网设备规模的不断扩大，设备管理系统应用需求也不断地增长。针对传统人工管理效率低的问题，提出融合语音技术与声纹认证的电力通信设备管理系统。首先，构建面向电力通信设备词汇的语音语料库和声学词表；然后，采用深度神经网络——隐马尔可夫模型（DNN-HMM）识别语音；最后，结合声纹认证技术，对操作人员的身份进行识别。该系统实现了从手动向语音智能交互操作的转变，使设备管理系统更加高效安全。
关键词：智能电网；设备管理系统；语音交互；声纹认证；语音语料库；声学词表；DNN-HMM
中图分类号：TN912.3 文献标志码：A 文章编号：1674-2605(2023)06-0008-06
DOI：10.3969/j.issn.1674-2605.2023.06.008
Power Communication Equipment Management System Integrating Voice Technology and Voiceprint Authentication
ZHENG Xuqi
(Jieyang Power Supply Bureau of Guangdong Power Grid Corporation, Jieyang 522000, China)
Abstract: With the continuous expansion of power grid equipment, the demand for equipment management system applications is also constantly increasing. Aiming at the problem of low efficiency in traditional manual management, a power communication equipment management system integrating voice technology and voiceprint authentication is proposed. Firstly, build a voice corpus and acoustic vocabulary for power communication equipment vocabulary; Then, deep neural network-hidden Markov model (DNN-HMM) is used to recognize voice; Finally, combined with voiceprint authentication technology, the operator's identity is identified. The system has achieved a transition from manual to voice intelligent interactive operations, making the device management system more efficient and secure.
Keywords: smart grid; equipment management system; voice interaction; voiceprint certification; voice corpus; acoustic vocabulary; DNN-HMM
