Real-Time Transcript and Participant Control in Zoom SDK App

Hi everyone,

I’m building a web application using the Zoom Meeting SDK (Embedded view). In my setup, three participants join each meeting:

  • One staff member (host)

  • A client

  • A candidate

The meeting is mainly between the client and candidate, and our platform supervises the session.

I’m trying to implement two things:

  1. Access real-time transcript text during the meeting at regular intervals so I can run my own AI-based analysis on the conversation.

  2. If my system detects PII (e.g., phone numbers, emails, etc.), I want to automatically mute participants and turn off their video programmatically.

I will handle the AI/PII detection logic on my side. My main questions are:

  • Does the Meeting SDK support access to live transcription text during an active meeting?

  • Is there a way to programmatically mute participants and disable their video using the SDK (assuming the app has host privileges)?

  • If supported, which APIs or SDK methods should I be looking at?

I’ve gone through the documentation but couldn’t find a clear way to achieve live transcript streaming or full participant media control from the embedded SDK.

Any guidance or clarification would be really appreciated.

Thanks in advance!

The Web Meeting SDK doesn’t provide a supported API to read live transcript text in‑app (it isn’t exposed “out of the box,” as confirmed by Zoom staff on the Dev Forum: Web SDK transcription limitation). If you need real‑time transcript data from Zoom Meetings, you should instead use Realtime Media Streams (RTMS) / the RTMS SDK, which is specifically designed to deliver live audio/video/transcript streams over a socket connection.

For participant audio control in Component View, you can mute or unmute participants (with host privileges) via EmbeddedClient.mute(mute, userId?). However, the Component View API reference for the EmbeddedClient namespace does not document any method to force‑stop another participant’s camera. Zoom staff have likewise confirmed on the Dev Forum that there currently isn’t an API to remotely disable video for a specific participant mid‑meeting.

If you’re looking for support integrating with Zoom RTMS, you can also take a look at Recall.ai (official Zoom RTMS partner). We’ve helped thousands of developers integrate with Zoom and can help you with your RTMS integration too!

Hi @amanda-recallai
Thank you for the detailed clarification — this really helps.

I appreciate you confirming the limitation around live transcription in the Web Meeting SDK and pointing me toward RTMS instead. That clears up the confusion I had after going through the documentation. The note about participant controls in Component View is also very helpful, especially the confirmation regarding video control not being supported.

Thanks as well for mentioning RTMS resources and the additional integration options — I’ll review those as I evaluate the best approach for our use case.

Really appreciate you taking the time to respond.