Have a live conversation about a basketball game with GPT4V, Whisper, TTS

Looks cool. But the delay is annoying. Not in this video only, I use voice commands quite a lot. Starting from Apple Watch to ChatGPT on iPhone. The delay is killing.

I can understand the delay in ChatGPT. But why translation from English to Spanish on Apple Watch takes so much time?

I think until we have instance reply from these machines, adoption will be growing slowly.

How on Earth did Steve Ballmer find his way into this demo?