Zoom RTMS - Tech Questions

Realtime Media Streams
,
1

  1. How can we clean up and send the audio and transcript info we have from the websocket calls, if suddenly the conversation got cut off?

  2. How can we seamlessly re-enable the connection once it is cut off and the user joins their meet again?

  3. In Otter AI, The AI itself joins as a user but, for us, we can only see our AI name at the bottom right of the user who is using our AI product

How can I allow our AI, to join as a user/participant itself?

Please answer these technical questions - or email me at the same email address provided in my profile.

The zoom rtms works really well in testing and looking forward to push it to production.

2

@gauthambh, I don’t know if question 1 and 2 in detail is asking about disconnection of websocket or disconnetion of user, but I’ll asner for both.

If a user disconnects from a meeting, it doesn’t affect RTMS, it will keep on running as long as the meeting is still running. Once the meeting ends, RTMS will get disconnected.

Scenario 1. RTMS websocket is disconnected due connectivity, you will need to initiate the reconnection. The sample apps and the SDK does incorporate reconnection logic in the scenario where it is disconnected unexpectedly.

Scenario 2. Host ends the meeting and starts another meeting, you will need RTMS to connect to this new meeting. This is typically easy to automate if you we using webhook to get notification that a meeting has started.

The question about Otter AI is not related to RTMS, and is more related to a Meeting SDK bot. I’m assuming you are asking how you can create your own AI meeting bot?

From a high level summary for AI meeting bot which joins as a participant, you will need a Linux, Windows or MacOS server, it will run many instances of Meeting SDK which meetings when it received notification that a meeting is running. You will need to work with raw audio and raw video in PCM and YUV420 format. The host would also need to explicitly give your bot permission before it can access the audio and video in meeting.

One of RTMS goal is to let developers easily access video/audio/transcript without going through the process of creating a meeting bot.

3

  1. I am using this guide - GitHub - zoom/zoomapps-advancedsample-react: This repository contains an Advanced Zoom Apps Sample. It should serve as a starting point for you to build and test your own Zoom App in development.

    I use the normal logic of connection with the websocket from our server rather than the zoom sdk.

    So you mean, I have to use the zoom rtms sdk for reconnection, if the connection dropped?

    1. Should I use the Meeting SDK to allow our AI to join as a user?

    Right now, the RTMS only scopes works well for our usecase.

    We grab only the raw audio in PCM format from the zoom servers in our websocket connection for our business usecase and it works well.

    Thanks.

@chunsiong.zoom

4

@gauthambh
No you do not need to use RTMS SDK. You have to handle the reconnection yourself if you are not using the SDK.

I cannot answer if you should use Meeting SDK or not as I do not know what is your requirement. If you are share more, I can provide some insights.

If you just want to grab the raw PCM audio for your requirement, RTMS is perfect.

5

@gauthambh here’s a list of samples you can refer to

6

Thanks @chunsiong.zoom, it was helpful.

We do not implement pentesting in our security flow, but implement SAST & DAST for every release we make.

When I tried to get open beta access, our app got rejected because of this.

Should we implement pen testing as a mandatory step for appstore approval.

Do reply.