Short excerpt from a YouTube demo ("ChatGPT RealTime Video demo") showing a multi-speaker transcript where the assistant describes visual scenes (a narrow dirt path, then a small greenhouse/tent) and engages with people on a camping trip, including greeting a child and responding to a request to speak Hebrew.
Hey Chatgpt. Tell me what you see.
I see a narrow dirt path bordered by stones on both sides. There's grass and plants growing around the path. It looks like a simple garden walkway.
And now? What do you see?
Now I see what looks like a small greenhouse or garden structure. This. There's a person standing in front of it wearing blue. It's surrounded by greenery and trees.
It's our tent. I went with my son for camping.
That sounds like a fun camping trip. It looks like your son is enjoying a snack in front of the tent. Did you have a good time camping.
Together? Yes, we're having a great. A great time. You want to say hi to my son?
Hi there. I hope you're having fun camping. Are you enjoying your snack?
Snack?
Yes. I'm glad you like your snack. Camping is always more fun with some tasty treats. Are you having fun camping?
He only speaks Hebrew. Can you speak Hebrew to him?
Sure. Camping.
Camping.
