Discussions

Hello,

I'm building a chatbot with custom RAG, video generation and custom voice. I have a few questions: What is the difference between sessions and video streaming minutes. Are the session audio synthesis, llm processing and video generation all in one? If so, then it costs less than purely using .speak method from agents SDK, which uses video streaming and costs 1 credit per 15 seconds, while sessions cost 0.12 per message?

This is more of a matter of whether to use my own llm setup, with my own voice gen from eleven labs and only rely on studioDID for video streaming?

Thanks a lot, I'm having a lot of issue understanding the pricing plans and subscription plan conditions