Have you done any looking around ? This is already done

I am not sure what you meant by that. I know that Vapi has voice to text and text to specch model encoding but even that can be avoided with direct voice to voice model with all remaining features of vapi. This is what I was looking for

@dl_amit - it's literally baked in and has an option for elevenlabs, cartesia and othes.

May be my point was not clear. I looked into the assistant setup in Vapi and we need to choose Model, transcriber and Voice tab respectively for their models. So this effectively means we are using speech to text and then text to speech conversion. Now Eleven labs came up with their voice agent setup as explained and I was wondering whether I can configure the voice agent and use from Vapi platform instead of configuring in Vapi

Hey @dl_amit , this isn't possible, but is there a reason for wanting to use the Vapi platform with an 11labs configured agent?

I found their vocie agent adhereing to instructions much better but then it cant connect to all mobile platform natively