A method, device and related equipment for acquiring meeting document information

Abstract

The present application discloses a method for obtaining conference document information, including collecting conference video information and conference audio information; performing feature extraction on the conference video information to obtain corresponding lip language features and text features; performing feature extraction on the conference audio information Extracting and obtaining corresponding audio features; performing feature fusion on the lip language features, the text features and the audio features to obtain fusion features; processing the fusion features to obtain meeting document information; obtaining the meeting document information The method can improve the comprehensiveness and accuracy of meeting records while ensuring the efficiency of meeting records. The application also discloses a meeting document information acquisition device, equipment and computer-readable storage medium, all of which have the above-mentioned beneficial effects.

Description

technical field [0001] The present application relates to the field of multimedia technologies, and in particular, to a method for obtaining conference document information, and also to an apparatus, device, and computer-readable storage medium for obtaining conference document information. Background technique [0002] In a video conference, it is particularly important to organize the video, audio, PPT and other information in the conference effectively and with high quality to form a complete written record. Traditional conference transcription methods mainly include manual recording through pen and paper and automated recording based on speech recognition technology. [0003] The method of manual recording by pen and paper is extremely inefficient; the automatic recording method based on speech recognition technology only obtains the transcribed document in a single voice mode, without other information in the meeting scene, such as the supplement of PPT information, etc...

AI Technical Summary

Problems solved by technology

[0003] The method of manual recording by pen and paper is extremely inefficient; the automatic recording method based on speech recognition technology only obtains the transcription document in a single voice mode, without other information in the meeting scene, such as the supplement of PPT information, etc. There is also the problem of incomplete information recording, and when the voice environment is relatively noisy, the accuracy of voice recognition will drop significantly, resulting in inaccurate meeting records

Images