Method, device, equipment and storage medium for determining a video conference speaker terminal

Abstract

The embodiment of the present invention discloses a method, device, device and storage medium for determining a video conference speaker terminal. The method includes: acquiring the audio level of the audio package from the terminal and the time stamp information of the audio package; according to the time stamp Information sequence extraction The audio package contained in the set time window is the target audio package; set the corresponding weight value for each target audio package according to the set rules, and determine the target audio level after multiplying the audio level of each target audio package by the corresponding weight value level; superimposing the target audio level, using the superposition result as the audio level of the terminal; determining that the terminal corresponding to the maximum audio level at the current moment is the video conference speaker terminal. The method of time window statistics is used to determine the speaker terminal of the video conference, and the size of the time window is automatically adjusted to make the switching of the speaker terminal more sensitive and stable.

Description

technical field [0001] The present invention relates to communication technology, in particular to a method, device, equipment and storage medium for determining a speaker terminal of a video conference. Background technique [0002] Video conferencing refers to a meeting in which people at two or more locations conduct face-to-face conversations through communication equipment and the Internet. In a video conference, participants can hear the voices of other conference sites, see the images, actions and expressions of other conference participants, and can also send electronic presentation content. [0003] There are often more than two terminals in a video conference, and the client often has a problem that the number of display windows is less than the number of terminals in the conference. In an actual video conference system, there is also a need to quickly shift the attention of the participants to the person who is speaking in the conference. Therefore, how to deter...

AI Technical Summary

Problems solved by technology

The terminal sends the status of the conference participants to the server according to a certain frequency, and the server judges the current conference speaker terminal, but the delay of several seconds will cause a relatively poor experience for the user; or the audio sampling data of each participant is counted in the preset The number of occurrences in the frequency band is used to judge the current conference speaker terminal to eliminate the noise of some specific frequency bands, which has high requirements for the environment. The sensitivity of switching the speaker terminal by this method is also relatively slow; or the terminal is based on the collected preset length threshold. The coherent voice signal of the current conference can be used to judge the current conference speaker terminal. When there are many users, it is very likely that the voice signal length of multiple terminals reaches the preset length at the same time, and it is difficult to determine the speaker terminal.

Images