Building with Gemini 2.0: Multimodal live streaming

Okay, this is a quick demo of multimodal live streaming in Gemini 2.0 and we are casting a tab into AI studio. Um, can you see my screen? Yes, I see your screen. It's showing a document with the title demo. It also contains bullet points and a Google Meet window with a person in the video. Is

there anything in particular you would like to talk about on the screen? Yeah, can you start the demo by just reading the text I highlighted? Sure, I can do that. The text you highlighted reads, "The multimodal live API lets you build real-time multimodal applications powered by Gemini 2.0 flash. It features tool use like Google

Search and code execution, function calling, and more." Would you like me to read the following lines, too? No, can you actually just tell me what this word means, though? I can do that. Multimodal refers to the ability to process and understand different types of data like text, images, and audio all together. Does

that help? Yeah, um I need to demo the interruptions feature next. Can you tell me a really boring kind of once upon a time sort of story? Okay, I can do that. Once upon a time, in a small village nestled beside a quiet forest, there was a very unremarkable rock. This rock was gray

and it was round Sorry, I have to interrupt you. No problem. I understand that interruptions are part of the demo. What should I do now? Um, so we're going to demo memory. Can you just summarize everything we've seen or heard so far? I can do that. So far, we've discussed a demo of the multimodal live API, which

lets you build real-time multimodal apps, and I define what multimodal means. I was also asked to tell a very boring once upon a time story, but I was interrupted. Is there anything else? Uh no, I think we're done. Um do you want to read this ending card or should I? I can read it if you would like. The