I had early access to GPT-6. This model will completely shatter your understanding of what's possible, and it will start a new era of creativity.
It's hard to explain how big of a jump this is, so I'll share my tests.
1/6 Made this 3D model and animation in code from an image.
@FelipeFumero663 Hey Felipe! The watch requests a websocket connection to the FastAPI server at the beginning of the conversation, if valid and authenticated, the server will start one with the OpenAI realtime endpoint
Testing some realtime voice with a live agent identity swap.
gpt-realtime-2 @OpenAI works very well also when switching from English to Italian and I used @opendot_ai for the control plane and webrtc connection.
Still need some tuning on the audio streaming part