Unlock instant, AI-driven research and patent intelligence for your innovation.

A speech extraction method for conference moderators based on speaker segmentation

A voice extraction and speaker technology, applied in voice analysis, instruments, etc., can solve the problems of large amount of calculation, many steps, difficult to realize the voice extraction of the conference host, etc., and achieve the effect of fast calculation speed and simple steps

Active Publication Date: 2016-08-10
SOUTH CHINA UNIV OF TECH
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Therefore, the current method has the disadvantages of many steps and a large amount of calculation, and it is difficult to achieve fast voice extraction of the conference moderator.

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
  • A speech extraction method for conference moderators based on speaker segmentation
  • A speech extraction method for conference moderators based on speaker segmentation
  • A speech extraction method for conference moderators based on speaker segmentation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0030] Such as figure 1 As shown, a method for extracting the voice of a conference moderator based on speaker segmentation includes the following steps:

[0031] S1. Read in an audio file recording the conference voice. The conference voice may be an audio file in various formats, such as WAV, RAM, MP3, VOX, etc.

[0032] S2. Use the voice detection method based on the threshold judgment to find the silent segment and the voice segment in the voice stream, splicing the above voice segments into a long voice segment in chronological order, and extract audio features from the long voice segment, and use the above extraction According to the Bayesian information criterion, the similarity between adjacent data windows in the long speech segment is judged to detect the speaker change point; finally, according to the above speaker change point, the audio file is divided into multiple speech segments, And each speech segment contains only one speaker, and the number of said speech ...

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 discloses a method for extracting the voice of a conference moderator based on speaker segmentation, comprising the following steps: S1, reading in an audio file with recorded conference voice; S2, speaker segmentation: detecting the speaker change point in the conference voice , taking the voice samples between two adjacent change points as a voice segment, and dividing the audio file into multiple voice segments; S3, voice segment distance comparison: the first voice segment after speaker segmentation is used as the conference host and compare the distance between the voice segment and other voice segments, and judge the voice segment whose distance is smaller than the threshold as the voice of the conference moderator, so as to obtain all the voice segments of the conference moderator. The invention lays a foundation for quick browsing of conference voices, topic extraction, speaker retrieval, etc., and has the advantages of being able to quickly and effectively extract the voice of the conference host.

Description

technical field [0001] The invention relates to speech signal processing and pattern recognition technology, in particular to a conference moderator speech extraction method based on speaker segmentation. Background technique [0002] The conference moderator refers to the speaker who makes the conference progress in an orderly manner in a multi-person conference. In frequent seminars, press conferences, speeches and other meetings, there is usually a meeting moderator. The moderator is often the first speaker of the entire meeting, he organizes and guides the participants to participate in the discussion of the meeting agenda in an orderly manner. Important information such as the theme of the meeting, the number and identity of the participants, the main agenda, and meeting resolutions can be obtained from the speech of the meeting host. This information is what people most want to obtain when browsing and analyzing conference voice. Therefore, extracting the voice of t...

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 Patents(China)
IPC IPC(8): G10L17/02
Inventor 李艳雄金海贺前华
Owner SOUTH CHINA UNIV OF TECH
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

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

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com