Voiceprint recognition method and system applied to electric power operation

• Main Authors: ZHOU SUJUN, HAN JING, ZHU MINGZENG, HUANG XINHUA, JIANG ZHIRU, LIU XIAOLAN, OU JIANMEI, SONG SIHUANG, WEN LIMING, HU KAIBO, LYU MING, LIANG CHAOCONG, MO ZIYING, MA HONGKANG, LIANG WEI, CHEN JIWAN, LI HEFENG, QIN QIUQIN
• Format: Patent
• Language: Chinese; English
• Published: 21.12.2021
• Subjects: ACOUSTICS; CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; MUSICAL INSTRUMENTS; PHYSICS; SPEECH ANALYSIS OR SYNTHESIS; SPEECH OR AUDIO CODING OR DECODING; SPEECH OR VOICE PROCESSING; SPEECH RECOGNITION

The invention discloses a voiceprint recognition method and system applied to electric power operation. The method comprises the following steps: acquiring voiceprint information of different people in an electric power operation scene; removing noise and interference information in the voiceprint information; performing non-negative matrix factorization (NMF) on the de-noised voiceprint information to extract features; acquiring a spectrogram of the voiceprint information based on the voiceprint information processed by the NMF; processing the spectrogram based on a convolutional neural network (CNN) voiceprint recognition algorithm; and outputting a voiceprint recognition result based on the convolutional neural network training model. According to the embodiment of the invention, the NMF is used for extracting the features in the feature extraction stage, and the multi-dimensional features are fused, so that a more accurate recognition effect can be obtained. 本发明公开了一种应用于电力作业中的声纹识别的方法及系统，其方法包括：采集电力作业场景中不同人的