Is there a way to dynamically change the transcriber, llm and voice during an active call? Thanks!!

@AiTitus :) Yes, you can dynamically change the transcriber, language model (LLM), and voice during an active call. This typically involves using an API that supports such operations, where you can send commands to switch between different transcribers, language models, or voices on-the-fly. However, the specific method to do this would depend on the capabilities of the API or platform you are using. If you're using VAPI, you would need to refer to the API documentation to understand how to implement these changes during a call. The API might provide specific endpoints or methods to update the call settings in real-time. If you need assistance with the specific commands or code to achieve this with VAPI, please provide more details about the current setup or the API documentation, and I'll be happy to help you with that!

Any ideas @Shubham Bajaj @Sahil

You can't do it right now

ahh ok maybe swap out with a dynamically generated squad in realtime? like create an assistant, add to the squad then dynamically transfer? thoughts @Sahil example?

@AiTitus :) as of know you cannot create and use

a dynamically generated squad in realtime

, you have to define the squads right before the call starts.

@AiTitus :) have you been able to solve this ? For example, to create a multilingual assistant, you'd create a squad with different transcriber - llm -voice, and the initial one would transfer to the right one, right ? I'm very curious how you'd set up the initial one then - probably using openai realtime as it is multilingual ?