Abnormal voice classifying method and device

Abstract

The embodiment of the invention provides an abnormal voice classifying method and device. The abnormal voice classifying method comprises the steps that framing processing is conducted on an abnormal voice signal to obtain N frames of signals, wherein N is an integer greater than 1; filtering processing is conducted on each signal in the N frames of signals according to a filter model to obtain energy feature information of the abnormal voice signal, wherein the filter model comprises M sub-filters, frequency intervals formed by the starting frequencies and stopping frequencies of the M sub-filters are continuous and are not overlapped mutually, different values of output energies of statistic Mel-spectra of multiple abnormal voice signal samples different in type on the M sub-filters are smaller energy threshold values, and M is an integer greater than 1. According to the energy feature information of the abnormal voice signal, an abnormal voice signal classifying result is determined. By adopting the abnormal voice classifying method and the abnormal voice classifying device, the abnormal voice classifying accuracy can be improved.

Description

technical field [0001] The embodiments of the present application relate to the field of audio signal processing, and more specifically, to a method and device for classifying abnormal sounds in the field of audio signal processing, and a method for establishing a filter model and the filter model. Background technique [0002] Public places refer to various places where the public engages in social life, such as squares, stations, schools, etc. All kinds of abnormal events and crimes in public places are closely related to people's life, property and safety. At present, the security monitoring of public places is mainly based on the video monitoring platform, and most of them do not have audio monitoring capabilities. Abnormal events in most public places are accompanied by abnormal sounds, such as gunshots, explosions, broken glass, screams, etc. The production of anomalous sounds indicates that anomalous events have occurred or are about to occur. Making up for the lac...

AI Technical Summary

Problems solved by technology

However, since abnormal sounds in public places include voice signals (such as screams) and non-speech signals (such as gunshots, explosions, and glass breaking), the fundamental frequency of abnormal sound signals in public places is scattered, and the signal energy varies between The frequency domain of the width is distributed intensively, and at the same time, it presents the characteristics of non-uniform distribution in the entire frequency domain. The above-mentioned traditional methods have limited description of the characteristics of abnormal sounds, resulting in poor classification results.

Images