Hey all , Is it possible to use gpt4.o api , I wanted to use its chat like audio chat human like speech! Lets say i have a user and the user should be able to speak to my gpt ( that i wanted to build ) and the gpt should speak like human
Until OpenAI releases their new “Advanced Voice Mode” in the API (hopefully sometime this year!), you have to build your own app yourself, that does Speech-To-Text to transcribe the user’s input, generates the response, then reads it out loud using Text-To-Speech. If you’re smart you can manage to lower the latency by using streaming.
If you want a quick and easy way to achieve this you can just build a custom GPT in ChatGPT and let users access it with the current Voice Mode, although you won’t have a lot of control on the voice. (Anyone correct me if I’m wrong, I’ve never built custom GPTs.)