Hey, the audio response of the text to speech mode just reads my prompt out loud instead of generating audio of what I’m describing it to generate. Is something wired up wrong under the hood here or is it supposed to work like that?
It is, it’s truly text to speech. The model select box are just different voices from OpenAI. It sounds like you are looking for text to audio which we haven’t implemented yet. We were holding off on certain AI stuff if it was actually useful, but Hugging Face has a number of Audio models we could add in. I don’t think OpenAI has general purpose audio. Have a great weekend.
I see, all good then. You have a great weekend too!