I have built a program where multiples ai interact, some process audio, some process images, and they all transcript everything to text. The problem its that the one taking care of everything, doesn’t recognize this and if a user ask (its fundamental for the purpose of the chatbot) it only respond that only process text and cant other even though in the irony that its responding that message trough audio, ive tried so many prompts and none of them works, i dont know if any of you could solve this, thank you very much!
1 Like
Welcome to the forum!
Just a thought.
Once the connection is made between the AI and a function to process the input, the connection is not closed and available for a new input. Thus I suspect a new connection needs to start by starting a new conversation or using an agent that can open and close connections, or have an agent to understand the name of the file and then redirect the file to a specific agent if that is possible.
3 Likes