Reputation: 11
I am looking to build a Microsoft Teams plugin for real-time voice translation. The requirements are:
The plugin should access the audio stream of a Microsoft Teams meeting.
Provide live transcription (speech to text) of the audio stream in the language of the speaker. For example - if speaker is speaking in English, provide live transcription in English.
Provide simultaneous live translation of the text. For example - provide live translation in Japanese.
I have an AI model to achieve steps 2 and 3 above. However, I don't know how to implement step 1 i.e. how to build a Teams plugin that could run my AI model in real time.
I have been exploring Microsoft Teams developer documentation (https://learn.microsoft.com/en-us/microsoftteams/platform/mstdd-landing) but there is so much to it and I am kind of lost.
For now, I am trying to create a real-time media bot using Microsoft Azure. This is still in progress - https://learn.microsoft.com/en-us/microsoftteams/platform/bots/calls-and-meetings/real-time-media-concepts https://dev.botframework.com/bots/new
It would be great if someone could share development experience as to how to go about building a Teams plugin for real-time voice translation. Thank you.
Upvotes: 0
Views: 322