Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Abnormal voice detecting method based on time-domain and frequency-domain analysis

A detection method and technology in the time-frequency domain, applied in speech analysis, speech recognition, instruments, etc., can solve problems such as inability to respond in time, inability to protect people's personal and property safety more effectively, and low level of intelligent early warning

Active Publication Date: 2012-09-12
NAT UNIV OF DEFENSE TECH
View PDF4 Cites 41 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

When an unsafe incident occurs, there are often abnormal voices such as calling for help, screaming, and shouting. The existing video surveillance system cannot respond to abnormal voices in a timely manner, and the level of intelligent early warning is not high, so it cannot protect the people more effectively. People's personal and property safety

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Abnormal voice detecting method based on time-domain and frequency-domain analysis
  • Abnormal voice detecting method based on time-domain and frequency-domain analysis
  • Abnormal voice detecting method based on time-domain and frequency-domain analysis

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0046] The abnormal human voice detection method designed in the present invention is mainly aimed at automatic monitoring of sound information in public places such as roads and streets or other key locations. Firstly, the abnormal segment is detected through time-domain energy difference, and then analyzed according to the frequency characteristics of the human voice. Whether the abnormal segment is an abnormal human voice, the specific process is as follows:

[0047] 1. Calculate the real-time background sound intensity of the monitoring scene, the specific process is:

[0048] Step1.1: Background sound intensity initialization

[0049] Continuously acquire L-segment equal-length short-term sound clips, calculate and average the intensity of each sound clip, and obtain the average sound intensity Initial intensity as background sound:

[0050] EL 0 ‾ = 1 L ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention relates an abnormal voice detecting method based on time-domain and frequency-domain analysis. The method includes computing the background sound intensity of a monitored scene updated in real time at first, and detecting and extracting suddenly changed fragments of the sound intensity; then extracting uniform filter Mel frequency cepstrum coefficients of the suddenly changed fragments; and finally using the extracted Mel frequency cepstrum coefficients of sound of the abnormal fragments as observation sequences, inputting a trained modified hidden Markov process model, and analyzing whether the abnormal fragments are abnormal voice or not according to frequency characteristics of voice. Time sequence correlation is improved when the hidden Markov process model is added. The method is combined with time-domain extraction of suddenly changed energy frames and verification within a frequency-domain range, the abnormal voice can be effectively detected, instantaneity is good, noise resistance is high, and robustness is fine.

Description

technical field [0001] The invention mainly relates to an abnormal human voice detection method based on time-frequency domain analysis. Background technique [0002] Safety prevention and control has increasingly become the focus of public attention, and video surveillance systems have been widely promoted and applied. Most cities above the county level and key places in towns and villages in my country have established video surveillance systems according to law to realize multi-directional and all-weather video surveillance and monitoring. storage. At present, most video surveillance systems do not monitor sound, or only do simple recording, lacking real-time analysis and processing of sound information. When an unsafe incident occurs, there are often abnormal voices such as calling for help, screaming, and shouting. The existing video surveillance system cannot respond to abnormal voices in a timely manner, and the level of intelligent early warning is not high, so it ca...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G10L15/10G10L15/14G10L19/00G10L19/025
Inventor 谢剑斌李沛秦刘通闫玮唐朝京谢昌颐
Owner NAT UNIV OF DEFENSE TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com