GPT-4o voice modality

Is there a way to use a GPT-4o text-to-speech capability?

My goal is to take a new text message as an input and return it as a sound file to a Telegramm messenger.
I have already done this successfully with the OpenAI TTS model. But I want a better solution with the GPT-4o. But I can’t find a way to get voice as output.

So, they didn’t released it yet: Will the API for the New Voice Be Released Separately? - #2 by PaulBellow - API - OpenAI Developer Forum

2 Likes