I am developing a bot that is a conversational AI, it keeps stats to enhance the conversation… it uses this to inform things like future style and emojis to use… however, these stats whilst useful in text, are very annoying to have read out in a Voice Chat.
Does anyone know if there’s a way to instruct the LLM to skip certain sections. e.g. generate this table, but never convert it to audio or read it out in voice chat?
Just don’t send that part to the API, maybe?
Welcome to the forum!
I’m really using just the GPT builder, I was wondering if there are some flags/comments that can be used, I’ve tried using a comment // “DO NOT READ OUT VIA VOICE CHAT” but I think this gets ignored due to the voice conversion being completely separate? Perhaps this needs to be posted in another section? sorry first time.
Yeah, I don’t think it’s possible the way you’re describing. Assistants maybe, though. Good luck.
That’s what I figured, voice chat is relatively new, so perhaps something along these lines might be implemented by OpenAI devs, I’ll live in hope for now. API might be a way to go, but it’s a lot of work to create a work around.