Discussions

I am looking to setup real time captions on the UI with liveavatar .

Based on this documentation https://docs.liveavatar.com/docs/full-mode-events#server-events there is an event avatar.transcription but this event is getting triggered after Heygen returns the whole response and the avatar completes it speech. We are using Heygen LLM to generate the response.

We want to get the transcription before the avatar starts speaking so that we play the words in the transcription as the avatar is speaking. Since we are getting the whole transcription at the end of the avatar speech we are not able to play it at the same time when the avatar speaks.

If there is any example on how to handle this , it would be appreciated