I’m excited about the potential performance gain using livekit inference. With plugins I got used to tweaking the parameters used for the STT and TTS (voice speed for example). How can I leverage those options when using livekit inference? The syntax in the examples seems to be limited to choosing a model provider and the voice.

Thank you for asking. Our intention is to let you pass in via extra_kwargs . However, we didn't prioritize getting all the params supported before launching. We're working on getting more provider paarameters supported. Which provider and which parameters would you like us to prioritize?

Cartesia TTS, Elevenlabs TTS, Deepgram STT, Google LLM

@crooked-dawn-90821 I'm curious which examples you were looking at and in general where you "looked first" to learn about how to configure the inference service. just trying to gain insight into how external people enter the documentation ecosystem!

I read the blog post announcement and followed the agent examples link at the end