The Design And Implementation Of Extended Function Modules Of Microsoft Audio And Video Conference System

During the COVID-19 pandemic,the demand for audio and video conferencing software in all walks of life shows an upward trend.For enterprises,employees need online meetings for team communication and collaboration.For schools,students need online meetings as online lectures.Besides,online collaboration can considerably improve work efficiency for enterprises.Also,schools might keep online lectures as a brand new teaching method in the future.In response to these requirements,Microsoft Teams team decides to upgrade the subsystem for audio and video calls in Teams,which supports one-on-one calls,group calls and meetings.Besides ensuring the quality of the call,the new subsystem for audio and video calls provides more functions that can considerably enhance user experience.The subsystem uses React to render UI components,uses Apollo Graph QL for front-end communication,and uses Azure Cognitive Services,Azure Bot Services and Azure Active Directory to provide services for the functions.This thesis mainly describes the requirements,design,implementation and testing of the new subsystem for audio and video calls,focusing on the persistence and multilanguage assistance related functions in calls.Based on the use cases,the subsystem is divided into four modules,namely: video recording module,audio transcript module,live translate module,and call history module.At present,the functions mentioned in this thesis have been completed and are in the internal trial stage.According to feedback,these new functions improve user experience mainly in two ways: facilitating the users to access the video and text record of the call,and make it easier to keep up with the speakers in multi-language calls.