URGENT - Stream 'audio.pcm' and 'video.yuv' in real time via websocket

I am working on the bot (a headless meeting bot using the Zoom Meeting Linux SDK within a Docker container) which joins the meeting, ask permission to record after which I am getting two files and ‘audio.pcm’ and ‘video.yuv’ after starting the recording. However, I want to process the audio and video in real time via websocket to wss://audio.info & wss://video.info respectively. Audio and Video should be extracted and sent in real-time for AI audio and video analysis. I want to process it using a neural network. How to do this effectively at production level? Detailed explanation along with some examples or repositories or code snippets etc would really help.

(Video websockets are optimized for those doing real-time AI video analysis, providing 720p PNG image frames at 2fps.)

@amanda-recallai
@ chunsiong.zoom
@ jake-recall-ai
@elliot-recallai

@pramodkoujalagi this is beyond the scope of this forum.

@pramodkoujalagi,

Another alternative to building your meeting bot is to use the Recall.ai API instead.
We power meeting bots for 400+ companies.

Here is a short demo: Zoom Bot Recording Demo | Loom

If you’re not able to get access, email me at amanda@recall.ai.

Let me know if you have any questions!