WEBVTT NOTE Sentence-level transcript of https://www.youtube.com/watch?v=xfPVBzDYjH4 NOTE One cue per sentence. Cue ids are the line anchors on /transcripts/xfPVBzDYjH4.html. A cue ends where the next begins, or 2 s after its last word. s1 00:00:12.480 --> 00:00:13.519 Can you hear me? s2 00:00:13.519 --> 00:00:14.880 Fantastic. s3 00:00:14.880 --> 00:00:16.080 Hello everyone. s4 00:00:16.080 --> 00:00:19.840 Well, hello couple people that wandered in. s5 00:00:19.840 --> 00:00:24.000 Um, I'm sorry the schedule has been changing around. s6 00:00:24.000 --> 00:00:29.439 I'm sorry you didn't find the quiet working area you were looking for. s7 00:00:29.439 --> 00:00:30.160 All right. s8 00:00:30.160 --> 00:00:36.079 Um, we are looking at some agents with Gemini API. s9 00:00:36.079 --> 00:00:41.280 Now, before we kick off, let's have a little listen to this. s10 00:00:41.280 --> 00:00:46.079 Welcome one and all to the grandest stage on Earth. s11 00:00:46.079 --> 00:00:52.480 You're standing right at the heart of the largest AI engineer World's Fair in history. s12 00:00:52.480 --> 00:00:56.719 Let's build the future together, shall we? s13 00:00:56.960 --> 00:01:01.280 Does anyone know what model that is? s14 00:01:01.680 --> 00:01:02.079 Yeah. s15 00:01:02.079 --> 00:01:04.080 You you you you're not allowed. s16 00:01:04.080 --> 00:01:05.040 Anyone else? s17 00:01:05.040 --> 00:01:07.840 Anyone know which model? s18 00:01:07.840 --> 00:01:11.119 No. Any guesses? s19 00:01:11.119 --> 00:01:18.799 It starts with G. Gemini. s20 00:01:18.799 --> 00:01:20.640 Yes, it is a Gemini model. s21 00:01:20.640 --> 00:01:23.520 But which one? s22 00:01:24.159 --> 00:01:25.439 No, it's not life. s23 00:01:25.439 --> 00:01:27.680 That's incorrect. s24 00:01:28.400 --> 00:01:28.799 Yes. s25 00:01:28.799 --> 00:01:30.880 This is Gemini TTS. s26 00:01:30.880 --> 00:01:32.640 Is that what you were about to say? s27 00:01:32.640 --> 00:01:33.759 No. Okay. s28 00:01:33.759 --> 00:01:36.479 Um, so Gemini text to speech. s29 00:01:36.479 --> 00:01:41.439 Uh, actually a really incredible model for expressive speech. s30 00:01:41.439 --> 00:01:44.320 You have inline audio text you can steer it with. s31 00:01:44.320 --> 00:01:46.000 Um, you can stream the speech. s32 00:01:46.000 --> 00:01:48.799 It's a lot lower latency now than it used to be. s33 00:01:48.799 --> 00:01:59.759 So uh if you are in the market for some speech uh text to speech that is Gemini TTS is a good model. s34 00:01:59.759 --> 00:02:04.960 Now let's have a look at uh some of the things we're doing today. s35 00:02:04.960 --> 00:02:07.520 Uh first of all my name is Thor. s36 00:02:07.520 --> 00:02:11.120 I work on the developer experience at Google DeepMind. s37 00:02:11.120 --> 00:02:16.160 So that mostly includes Gemini API and Google AI Studio. s38 00:02:16.160 --> 00:02:23.040 So if you're using the Gemini models through the API, uh you know, let me know if you have any feedback. s39 00:02:23.040 --> 00:02:28.000 I have my colleague Philip here as well, uh who's working on managed agents on the API. s40 00:02:28.000 --> 00:02:32.640 So if you're using that, uh let us know as well. s41 00:02:32.640 --> 00:02:38.400 So this is a bit of a a prelook at the workshop we're doing tomorrow. s42 00:02:38.400 --> 00:02:55.040 Um so tomorrow we'll do kind of Gemini API especially focused on um live agents uh conversational agents real time uh conversational agents multimodal real-time conversational agents we're just packing everything in there s43 00:02:55.040 --> 00:03:04.400 um and so what we can look at actually there's uh been a lot of things that shipped and so specifically what we want to look at is the multimodality s44 00:03:04.400 --> 00:03:17.599 of Gemini and the gen media models Uh so actually this slide is already out to date because this morning we shipped a new model uh that is called anyone know which model did we ship? s45 00:03:17.599 --> 00:03:26.000 We shipped Omni Flash which is uh a video generation and editing model and we shipped Nano Banana Light. s46 00:03:26.000 --> 00:03:35.680 So Nano Banana Light is a really exciting model because it cuts down the latency of image generation and image editing quite significantly. s47 00:03:35.680 --> 00:03:39.040 And so what I've done, we'll we'll do a couple demos here. s48 00:03:39.040 --> 00:03:48.720 Uh and one of the demos is um this example here, which I call the uh Gemini Nanobanana story canvas. s49 00:03:48.720 --> 00:03:55.200 Uh and I just VIP code this up uh just recently and I'm having some trouble with my mouse here. s50 00:03:55.200 --> 00:04:07.200 Um but so what we can do is we can come up with a story with um Gemini Life and then have it use Nano Banana Light to to illustrate that story for us. s51 00:04:07.200 --> 00:04:10.720 So anyone you know this is a little bit like improv theater now. s52 00:04:10.720 --> 00:04:15.120 Anyone has a story line they want to start with? s53 00:04:15.840 --> 00:04:18.799 Where are we somewhere? s54 00:04:18.799 --> 00:04:26.720 Okay, let's say um we're in Taipei in the 1980s. s55 00:04:27.120 --> 00:04:36.080 The neon lights of Taipe's night market pulsed with an almost magical energy reflecting off the wet pavement and the chrome of passing scooters. s56 00:04:36.080 --> 00:04:40.000 Rain had just passed, leaving a humid mist that hung in the air. s57 00:04:40.000 --> 00:04:48.000 Amidst the throng of shoppers and vendors, a young woman clutched a small, strangely glowing object under her jacket. s58 00:04:48.000 --> 00:04:51.680 Something very exciting happened in 1980s Taipei. s59 00:04:51.680 --> 00:04:54.720 A new kind of drink emerged. s60 00:04:54.800 --> 00:05:01.199 She paused at a stall where a man was vigorously shaking a metal cup, creating a frothy, milky concoction. s61 00:05:01.199 --> 00:05:08.479 He handed a cup to a customer who took a long sip with wide eyes, savoring the strange tapioca pearls at the bottom. s62 00:05:08.479 --> 00:05:14.639 This was the birth of bubble tea, and its sweet, chewy magic was spreading like wildfire. s63 00:05:18.000 --> 00:05:21.520 So what you can see here is uh from the tool call. s64 00:05:21.520 --> 00:05:31.520 So Gemini Life is getting a tool call um and then it's using the Gemini interactions API to generate an image on the fly. s65 00:05:31.520 --> 00:05:40.479 Now it still takes kind of you know two to three seconds for the actual image generation but that is um pretty impressively fast. s66 00:05:40.479 --> 00:05:46.160 And um what you can look at is if you're looking at the code here uh we're using the interactions API. s67 00:05:46.160 --> 00:05:55.440 And so the interactions API is exciting because it is a stateful API which means we can chain our interactions very seamlessly. s68 00:05:55.440 --> 00:06:09.039 So you could see here the first image generation that we were doing uh it didn't have a previous interaction ID but then the next generation that we were doing we're passing this previous interaction ID s69 00:06:09.039 --> 00:06:18.960 to the interactions API which means that we have the previous context of the previous interaction and we can continue that in the generation. s70 00:06:18.960 --> 00:06:30.880 So what that means specifically for our demo here is that you can see is our first you know interaction uh it generated an image with a very specific kind of visual feeling and also with s71 00:06:30.880 --> 00:06:38.960 characters right and so we can continue that interaction and so we can keep the same consistency kind of across these generations. s72 00:06:38.960 --> 00:06:48.560 So if you haven't tried out the interactions API, uh it is our unified API for interacting with all of the Google DeepMind models as well as agents. s73 00:06:48.560 --> 00:06:50.720 Uh and so do try that out. s74 00:06:50.720 --> 00:06:57.759 It is a it is a really really cool um way to interact with everything that Google DeepMind has. s75 00:06:57.759 --> 00:07:01.759 Okay, so this was you know a bunch of things that happened. s76 00:07:01.759 --> 00:07:04.319 Um so this was kind of on the Gen Media side. s77 00:07:04.319 --> 00:07:07.520 We shipped uh Nano Banana Light this morning. s78 00:07:07.520 --> 00:07:12.720 Now, we also recently shipped on the audio side our live translation model. s79 00:07:12.720 --> 00:07:17.440 Has anyone tried out Gemini 3.5 live translate? s80 00:07:18.160 --> 00:07:18.479 Yes. s81 00:07:18.479 --> 00:07:19.919 Okay, a couple folks. s82 00:07:19.919 --> 00:07:21.039 Uh, that's fantastic. s83 00:07:21.039 --> 00:07:24.560 So, what we can do is uh we can try that out as well. s84 00:07:24.560 --> 00:07:27.039 We can play uh a little demo. s85 00:07:27.039 --> 00:07:39.199 And so what I'm going to do here is I'm going to create this live translation session and I want you to scan uh this QR code on your phone. s86 00:07:39.199 --> 00:07:45.440 And so what this is doing here is if I now enable um my microphone. s87 00:07:45.440 --> 00:07:51.759 So this is getting the the microphone of my speech um here into the session. s88 00:07:51.759 --> 00:07:56.639 And now if you go in and you select your preferred language. s89 00:07:56.639 --> 00:07:58.639 So we're seeing here we're getting some German. s90 00:07:58.639 --> 00:08:02.080 So someone is listening to me now speaking in German. s91 00:08:02.080 --> 00:08:06.960 And so if you are uh now listening to this, so you probably need some headphones. s92 00:08:06.960 --> 00:08:14.879 So if you have some noiseancelling headphones, you can now listen into this talk in your preferred language. s93 00:08:14.879 --> 00:08:18.560 And so Gemini live translate is a really incredible model. s94 00:08:18.560 --> 00:08:20.479 It's near realtime translation. s95 00:08:20.479 --> 00:08:33.279 So as I'm speaking, as soon as the model is recognizing kind of the context of what I'm speaking about, uh it will then start translating in real time to your desired language. s96 00:08:33.279 --> 00:08:35.120 So I see someone is listening in there. s97 00:08:35.120 --> 00:08:36.640 Do do you get some audio out? s98 00:08:36.640 --> 00:08:38.080 Can you hear it? s99 00:08:38.080 --> 00:08:38.320 Yeah. s100 00:08:38.320 --> 00:08:39.760 Does it sound good? s101 00:08:39.760 --> 00:08:40.560 Fantastic. s102 00:08:40.560 --> 00:08:52.880 See, um, so this is this is really exciting because you know in a conference like this obviously we have a very very international audience and so I can speak in my preferred language. s103 00:08:52.880 --> 00:08:56.720 So I could switch to German now for example. s104 00:09:11.200 --> 00:09:14.880 the powerful Gemini live translates. s105 00:09:14.880 --> 00:09:17.440 Um, it is a a really incredible model. s106 00:09:17.440 --> 00:09:26.240 Um, it automatically detects the language the speaker is speaking and then translates it in real time uh into our preferred language. s107 00:09:26.240 --> 00:09:30.160 Okay, we have uh a very international crowd here today. s108 00:09:30.160 --> 00:09:33.839 Uh, who's listening in in uh Chinese? s109 00:09:33.839 --> 00:09:36.720 Are you listening in Chinese? s110 00:09:37.040 --> 00:09:38.000 Oh, okay. s111 00:09:38.000 --> 00:09:44.240 I didn't understand that, but okay. s112 00:09:44.240 --> 00:09:48.560 Um, so that is Gemini live translate. s113 00:09:48.560 --> 00:09:54.480 Um, if you want to try it out, you can add it to your um applications now. s114 00:09:54.480 --> 00:09:56.959 It is available on the Gemini API. s115 00:09:56.959 --> 00:09:58.640 So, do try that out. s116 00:09:58.640 --> 00:10:04.959 uh it is a really incredible model and uh would love to hear kind of what you're what you're building with that. s117 00:10:04.959 --> 00:10:09.200 Now obviously we're looking at uh agents, right? s118 00:10:09.200 --> 00:10:12.640 Can your agents uh what's it the title of this talk? s119 00:10:12.640 --> 00:10:14.160 Gemini made up this title. s120 00:10:14.160 --> 00:10:16.640 So I uh can your agent hear me now? s121 00:10:16.640 --> 00:10:18.800 I think that that is the title. s122 00:10:18.800 --> 00:10:26.959 So our most recent real time conversational model is called Gemini 3.1 Flash Life. s123 00:10:26.959 --> 00:10:34.480 Now, if you've talked to Gemini before in the Gemini app, that is exactly the model that you're speaking to. s124 00:10:34.480 --> 00:10:36.399 And so, we can do a little demo here. s125 00:10:36.399 --> 00:10:41.680 So, this demo was vipcoded by Gemini in Google AI Studio. s126 00:10:41.680 --> 00:10:54.000 And if you remember the good old times, you know, when you were listening to radio, you know, not too long ago, you could phone up the the the radio host and you could ask for a song, right? s127 00:10:54.000 --> 00:10:55.920 You could wish for a song. s128 00:10:55.920 --> 00:10:57.600 So, it's the same idea here. s129 00:10:57.600 --> 00:11:11.519 We're using Gemini Life to be our radio DJ and then we're giving Gemini Life an API, a function call to the Gemini API to then use LIA 3, which is our music generation model s130 00:11:11.519 --> 00:11:14.959 to then generate some music for us on demand. s131 00:11:14.959 --> 00:11:17.440 Okay, let's uh try this out. s132 00:11:17.440 --> 00:11:20.399 We're we're just going live demo to live demo here. s133 00:11:20.399 --> 00:11:22.160 Let's let's see. s134 00:11:22.160 --> 00:11:23.440 Well, hello there. s135 00:11:23.440 --> 00:11:25.920 You're live on the live jukebox right now. s136 00:11:25.920 --> 00:11:28.959 What absolute banger are we cooking up for you today, mate? s137 00:11:28.959 --> 00:11:32.480 To get this track properly sorted, I need a bit more of an idea. s138 00:11:32.480 --> 00:11:33.760 What genre are we thinking? s139 00:11:33.760 --> 00:11:37.120 Upbeat synth pop, smooth jazz, something with a bit more well. s140 00:11:37.120 --> 00:11:38.880 And what's the vibe lyrics-wise? s141 00:11:38.880 --> 00:11:39.920 Give us some details. s142 00:11:39.920 --> 00:11:40.160 Yeah. s143 00:11:40.160 --> 00:11:44.320 Can we get like a really high energy ragaton, please? s144 00:11:44.320 --> 00:11:50.000 About the AI engineering world's fair here in San Francisco. s145 00:11:50.160 --> 00:11:52.560 Now that sounds like a proper belter, mate. s146 00:11:52.560 --> 00:11:56.000 high energy regaton about the AI engineering world's fair. s147 00:11:56.000 --> 00:11:56.720 Brilliant. s148 00:11:56.720 --> 00:11:56.959 Right. s149 00:11:56.959 --> 00:12:05.200 To make this absolutely top of the pops, any specific bits of tech or big names you want thrown in, or are we just going for that huge festival vibe? s150 00:12:05.200 --> 00:12:08.560 Yeah, just surprise me, please. s151 00:12:09.120 --> 00:12:10.720 You got it, my friend. s152 00:12:10.720 --> 00:12:14.880 Prepare for a huge musical surprise generating that track right now. s153 00:12:14.880 --> 00:12:19.200 Mixing that regaton rhythm with the tech buzz for the World's Fair. s154 00:12:19.200 --> 00:12:21.519 It's going to blow your headphones off. s155 00:12:21.519 --> 00:12:21.839 Fair. s156 00:12:21.839 --> 00:12:22.480 What a feeling. s157 00:12:22.480 --> 00:12:25.519 Let's go on the floor. s158 00:12:25.519 --> 00:12:26.800 Neural networks firing up. s159 00:12:26.800 --> 00:12:28.000 Groundbreaking for sure. s160 00:12:28.000 --> 00:12:30.560 Clever algorithms dancing. s161 00:12:30.560 --> 00:12:31.839 Tech pioneers mingling. s162 00:12:31.839 --> 00:12:32.959 This is innovation. s163 00:12:32.959 --> 00:12:34.320 Baseline is pulsing heavy. s164 00:12:34.320 --> 00:12:35.440 Can you feel the heat? s165 00:12:35.440 --> 00:12:36.880 Sin stabs electrifying. s166 00:12:36.880 --> 00:12:39.440 Move in your feet. s167 00:12:41.839 --> 00:12:44.480 This is how we are. s168 00:12:50.079 --> 00:12:51.519 Yeah, that that was pretty cool. s169 00:12:51.519 --> 00:12:54.079 You went multilingual there as well. s170 00:12:54.079 --> 00:12:54.560 Nice one. s171 00:12:54.560 --> 00:12:56.560 Cheers. s172 00:12:57.040 --> 00:12:57.839 Cheers, mate. s173 00:12:57.839 --> 00:12:59.200 Glad you like the flavor. s174 00:12:59.200 --> 00:13:00.639 Keep those big ideas coming. s175 00:13:00.639 --> 00:13:02.079 This jukebox never sleeps. s176 00:13:02.079 --> 00:13:03.360 What are we spinning next then? s177 00:13:03.360 --> 00:13:04.800 Feeling. s178 00:13:04.800 --> 00:13:05.360 All right. s179 00:13:05.360 --> 00:13:08.320 Um, yes, you can try this out yourself. s180 00:13:08.320 --> 00:13:13.519 Uh, it is is a a demo that was built in Google AI Studio. s181 00:13:13.519 --> 00:13:19.040 Uh so if you want you can fork this in AI studio and continue yourself. s182 00:13:19.040 --> 00:13:30.399 Um I know this application might not necessarily be that useful for your business context, you know, if that's something you're looking to do, but I think the individual components s183 00:13:30.399 --> 00:13:31.680 are definitely useful. s184 00:13:31.680 --> 00:13:41.360 So if you're building real-time conversational agents, uh you know, you want them to be multilingual, you want them to have capabilities, uh tool calls. s185 00:13:41.360 --> 00:13:49.040 So in this example, we were just giving the agent a tool call to generate music with LIA 3 uh on demand. s186 00:13:49.040 --> 00:13:58.240 Now obviously this could also just call out to your CRM and get some you know customer information or it could create you know a new ticket on the back end. s187 00:13:58.240 --> 00:14:07.519 Um so you know really what I want you to take away is kind of this real time multimodal conversational uh experience that you can build. s188 00:14:07.519 --> 00:14:18.959 And so the way that you can build with this model, so the the Gemini 3.1 flash real-time conversational model is through the life API. s189 00:14:18.959 --> 00:14:25.519 So the life API is a low latency real time voice and vision uh API. s190 00:14:25.519 --> 00:14:30.720 So you can build multiodal uh life real-time interactions. s191 00:14:30.720 --> 00:14:45.519 And so what this means is you have this stateful websocket API where you create a session and then over this websocket you can send in real time text, audio and video frames to the model s192 00:14:45.519 --> 00:14:50.320 and the model will then respond in real time back in audio chunks. s193 00:14:50.320 --> 00:14:54.639 And these audio chunks you can then stream uh in your application. s194 00:14:54.639 --> 00:15:00.560 Uh and in addition to the audio, you also get the transcript of what was said. s195 00:15:00.560 --> 00:15:11.040 So if you've worked with voice agents before, this is a little bit different from kind of your cascading voice agent pipeline because this is a native audio model. s196 00:15:11.040 --> 00:15:19.120 So the model itself does not go through text, but rather the intelligence is baked into the audio model. s197 00:15:19.120 --> 00:15:30.160 So the audio model can understand the speech naturally and natively and can then respond back in any of the 90 supported languages. s198 00:15:30.160 --> 00:15:36.320 So this is what allows us to to switch very fluidly between different languages. s199 00:15:36.320 --> 00:15:40.399 Um so kind of the key features here, yeah, multilingual support. s200 00:15:40.399 --> 00:15:44.320 So 90 uh different languages currently supported in the preview. s201 00:15:44.320 --> 00:15:47.120 Um we can barge in, we can interrupt the model, right? s202 00:15:47.120 --> 00:15:57.680 If our DJ is talking a bit too lengthy, uh we we can we can barge in, interrupt it, we can give it uh the tools and we have the audio transcription as well. s203 00:15:57.680 --> 00:16:09.120 Now, if you're building real time conversational applications or real-time um multimodal applications, uh you know, real-time media can be quite challenging to scale. s204 00:16:09.120 --> 00:16:11.759 So, we have a great lineup of partners. s205 00:16:11.759 --> 00:16:16.240 So, you know, LifeKit, Pipecat, they all actually here in the expo as well. s206 00:16:16.240 --> 00:16:17.839 I recommend you go visit them. s207 00:16:17.839 --> 00:16:25.120 They do some really really cool stuff to allow you to get up and running quickly with the Gemini Life API. s208 00:16:25.120 --> 00:16:30.639 Okay, so if you're now um you know, you're like, "Okay, this sounds great. s209 00:16:30.639 --> 00:16:31.519 I want this." s210 00:16:31.519 --> 00:16:36.959 Um the easiest way to try this out is through Google AI Studio. s211 00:16:36.959 --> 00:16:40.880 So we spend a lot of money on this domain, AI.studio. s212 00:16:40.880 --> 00:16:43.519 So I do appreciate if you use that one. s213 00:16:43.519 --> 00:16:45.519 Um so we can get some ROI there. s214 00:16:45.519 --> 00:16:47.759 So ai.studioife. s215 00:16:48.160 --> 00:16:49.440 You can try out the model. s216 00:16:49.440 --> 00:16:54.720 You can talk to it and you can understand the capabilities that are available. s217 00:16:54.720 --> 00:16:56.560 So let's do that now. s218 00:16:56.560 --> 00:16:59.519 Uh so we're just going to AI.dev. s219 00:16:59.519 --> 00:17:01.440 We also have AI.dev. s220 00:17:01.440 --> 00:17:03.839 Um yeah, please do use that as well. s221 00:17:03.839 --> 00:17:07.039 That one was was also not not so cheap. s222 00:17:07.039 --> 00:17:08.160 Uh there we go. s223 00:17:08.160 --> 00:17:10.160 AI.devlife. s224 00:17:10.720 --> 00:17:15.760 And so here uh we can also try out the live translate model if you want to. s225 00:17:15.760 --> 00:17:21.679 Uh but we can here select our Gemini 3.1 flash uh live model. s226 00:17:21.679 --> 00:17:23.520 We can select a different voice. s227 00:17:23.520 --> 00:17:25.760 Uh for example, I like puck here. s228 00:17:25.760 --> 00:17:29.919 And then also we can enable grounding with Google search. s229 00:17:29.919 --> 00:17:37.120 So obviously the model is trained on a lot of information but it does not have access to real time information, right? s230 00:17:37.120 --> 00:17:45.520 And so by giving it access to Google search, it can actually go off and it's basically like a built-in function call. s231 00:17:45.520 --> 00:17:56.000 And if we ask it about today's weather or you know why Germany played so well in the World Cup, uh we can we can do that and it can use Google to to look that up. s232 00:17:56.000 --> 00:17:58.880 Okay, let's give this a try. s233 00:17:58.880 --> 00:18:01.440 Hey, can you hear me? s234 00:18:01.440 --> 00:18:03.440 Yes, I can hear you perfectly. s235 00:18:03.440 --> 00:18:04.880 How can I help you today? s236 00:18:04.880 --> 00:18:07.200 Can you see me? s237 00:18:07.200 --> 00:18:08.960 Yes, I can see you clearly. s238 00:18:08.960 --> 00:18:13.440 You're wearing a black shirt and a black hat sitting in front of a blue background. s239 00:18:13.440 --> 00:18:15.200 Anything specific you wanted to show me? s240 00:18:15.200 --> 00:18:15.440 Yeah. s241 00:18:15.440 --> 00:18:18.400 What does it say on my sweatshirt? s242 00:18:18.400 --> 00:18:21.840 I can see the word Gemini printed in yellow letters. s243 00:18:21.840 --> 00:18:25.039 Anything else you'd like me to identify? s244 00:18:44.720 --> 00:18:45.919 Uh, okay. s245 00:18:45.919 --> 00:18:50.160 I think my uh my Mandarin didn't work out so well there. s246 00:18:50.160 --> 00:18:53.360 Uh, can you also speak Mandarin, please? s247 00:18:53.360 --> 00:18:55.360 Yes, I can speak Mandarin, too. s248 00:18:55.360 --> 00:18:58.320 What would you like to say now? s249 00:19:06.480 --> 00:19:08.799 What's about? s250 00:19:13.200 --> 00:19:14.000 All right. s251 00:19:14.000 --> 00:19:16.960 If you don't read Mandarin, I do apologize. s252 00:19:16.960 --> 00:19:23.679 I was just asking the model what it has eaten, and I just said it doesn't need to eat because it's it's a model. s253 00:19:23.679 --> 00:19:27.840 So, uh, yeah, there we go. s254 00:19:27.840 --> 00:19:28.559 All right. s255 00:19:28.559 --> 00:19:36.799 So this is the easiest way for you to get started with uh Gemini ai.studiolife. s256 00:19:37.679 --> 00:19:49.440 If you want to learn more about building with Gemini and the Gemini live API, join me tomorrow morning in uh room 2009, I think. s257 00:19:49.440 --> 00:19:50.400 Check the schedule. s258 00:19:50.400 --> 00:19:52.960 It it does change every once in a while. s259 00:19:52.960 --> 00:19:56.640 Uh but other than that, Dankur, thank you. s260 00:19:56.640 --> 00:19:58.640 Cheers.