A microphone voice collection method in a multi-person conference scenario

By separating and analyzing audio signals in a multi-person conference scenario, calculating the probability index of the main speech components, and adjusting the weight reconstruction signal, the problems of low quality of speech signal reconstruction and low processing timeliness in the prior art are solved, and a higher quality audio processing effect is achieved.

CN119479678BActive Publication Date: 2025-03-25GUANGZHOU AISUO ELECTRONIC PROD CO LTD
2 Cites 0 Cited by

Patent Information

Application Number
CN202510031305.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-01-09
Publication Date
2025-03-25
Estimated Expiration
2045-01-09

AI Technical Summary

Technical Problem

The prior art is difficult to effectively utilize audio phase information in multi-person conference scenarios, resulting in low quality of voice signal reconstruction and low processing methods, which affects real-time effects.

Method used

By obtaining the sound source audio captured by each microphone in a multi-person conference scenario for pre-processing, the original component signal is separated, the signal possibility is analyzed, the possibility index of the main speech component is calculated, the weight reconstructed signal is adjusted, and the reconstructed audio information is obtained according to the signal strength.

Benefits of technology

The quality of the reconstructed signal is significantly improved, and the overall effect of audio processing is improved, allowing remote participants or conference recording devices to capture and understand conference content more clearly, enhance user experience, and maintain high-quality signal processing effects in complex acoustic environments.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The present invention relates to the field of speech recognition technology, and in particular to a microphone speech collection method in a multi-person conference scenario, comprising: obtaining sound source audio captured by any microphone in the multi-person conference scenario and performing preprocessing to obtain sound source data to be analyzed; separating into multiple original component signals, analyzing the possibility that the original component signal is a main signal to obtain a main signal group, calculating the possibility index of the main speech component based on a target analysis signal, and obtaining the possibility that the target analysis signal is the main component information corresponding to the microphone of the main speaker; then obtaining the final possibility that the target analysis signal is the main component information corresponding to the microphone of the main speaker; determining an adjustment factor based on the final possibility of the main component information corresponding to the microphone of the main speaker, and reconstructing each corresponding signal by weighting the adjustment factor; calculating the signal strength of any microphone in combination with the reconstructed signal, and obtaining the reconstructed audio information of the corresponding microphone by the signal strength.
Need to check novelty before this filing date? Find Prior Art