How to use Zoom AI Companion Summary with UI Toolkit (Video SDK)?

Hi Zoom Team,

I’m using the Zoom UI Toolkit for Web (which uses the Video SDK) to build a custom video experience.

I’m interested in leveraging Zoom AI Companion features — especially the Meeting Summary — within this app. However, from my research, it looks like these features are only supported in the Meeting SDK or the Zoom desktop/web client, not in the Video SDK.

Could you please clarify the following:

  1. Can I use AI Companion Summary with the UI Toolkit / Video SDK?
  2. If not, is there a recommended approach or workaround (like combining the Meeting SDK with the Video SDK)?
  3. Are there any future plans to extend AI Companion features to the Video SDK?

Thanks for your time and support!

Best regards,

Hi @jeff9 let me ask our video sdk team if this feature is available today. I will update you shortly.

@ticorrian.heard Please let me know if there is any update on this one ?

Hi @jeff9 apologies for the wait. At the moment, we do not support AI Companion in our Video SDK. I am still waiting for my engineering team on any updates for a roadmap though. I’ll update you as soon as they get that answered.

Hi @ticorrian.heard

Thanks for the update — no problem at all. I understand that AI Companion isn’t currently supported in the Video SDK. I’ll keep an eye out for any updates from the engineering team regarding the roadmap. Appreciate you keeping us in the loop!

Best,
@jeff9

Hi @ticorrian.heard,

Thanks again for the update.

In the meantime, I’ve been experimenting with a workaround: I’m recording the meeting audio, then using a transcription service to generate a transcript. From there, I use ChatGPT to create meeting notes based on the transcript.

This approach works to some extent, but the key limitation is that the transcript doesn’t identify who is speaking. Without speaker attribution, the generated summaries lack context and accuracy.

I noticed that the Video SDK UI Toolkit does support transcription, but I haven’t been able to get it working yet. From what I understand, transcription requires a paid plan — we’re currently on the Video SDK | Pay As You Go plan, but it’s unclear whether that includes access to transcription features or if we need to upgrade.

Could you clarify:

  • Does the Pay As You Go plan support transcription with speaker names?
  • If not, which plan would we need to enable that?

Appreciate your help and guidance as always!

Best regards,
@jeff9

Hi @jeff9 thats an interesting use case. For our transcription service, we do offer live transcription and callback events in the SDK where you can better automate this without waiting for a 3rd party service. The data given in the payload includes who the speaker is as well as a timestamp for the event. Transcription is a separated license which would be added to your account by our sales team.

Have you tried contacting them?