Loading video...

Video Failed to Load

There was a problem loading this video. This could be due to a temporary network issue or the video might be unavailable.

I’ve been exploring Gemini 2.0’s new native audio output capability, which is available for early testers. I’m a developer at Google Creative Lab, and wanted to share one of my favorite experiments so far called ✨ VoiceCursor (🔊 sound on for video) Unlike traditional TTS, native audio lets you... show more

Trudy Painter

3,433 subscribers

67,672 views • 1 year ago •via X (Twitter)

Education Science & Technology

Anya Rossi• Live Now

Private livecam show

10 Comments

Trudy Painter1 year ago

Gemini 2.0 native audio output is available in AI Studio for early testers. The prompt in this screencap is: Say this in an upbeat, happy tone: “You can steer a voice and … put emphasis on different words!” 🔗

Trudy Painter1 year ago

✨Voice Cursor follows a similar prompting strategy. After you highlight a phrase, the Voice Cursor will ask the API for audio for the phrase in your selected voice and tone. (and you can edit the prompt sent to the Gemini API in the bottom box)

Trudy Painter1 year ago

And for me, when the ✨Voice Cursor sits inside a familiar text editor, the highlight interaction feels fluid and comfortable. I’m excited about how native audio might enable new kinds of tools for how we write...

Trudy Painter1 year ago

You can get the code to see how it works at Native audio output is available to early testers now, with a wider rollout expected next year. This voice cursor was built on top of Such a good repo Also - it’s super simple to change the tone prompt presets + how you make calls to the Gemini 2.0 API (see screenshot below).

jpa1 year ago

so cool, trudy!

Codetard1 year ago

:)

Tom Bielecki1 year ago

@codexeditor audio as another annotation layer

Data & Analytics1 year ago

@JeffDean @JeffDean, that native audio output sounds dope! Real game-changer for developers. How’s it stacking up against other tools you’ve tried?

steve ike1 year ago

This is really cool. Thanks for sharing, look forward to checking out the code and learning from your work.

𝑫𝒂𝒏𝒊𝒆𝒍 𝑺𝒄𝒐𝒕𝒕 𝑴𝒂𝒕𝒕𝒉𝒆𝒘𝒔 🇦🇺1 year ago

Oh, the quality is remarkable! Thanks for sharing.

Related Videos

🚨 Breaking: Google is rolling out a native image feature in AI Studio! You can now use Gemini 2.0 Flash Experimental for native image output. This feature is insanely powerful , you can do so much with it. Google is crushing it, no hype, just delivering. Mad respect for the Google DeepMind and Gemini team! I had early access to this for a long time but I wasn’t supposed to share it :) It's amazing, you guys will love it.

🚨 Breaking: Google is rolling out a native image feature in AI Studio! You can now use Gemini 2.0 Flash Experimental for native image output. This feature is insanely powerful , you can do so much with it. Google is crushing it, no hype, just delivering. Mad respect for the Google DeepMind and Gemini team! I had early access to this for a long time but I wasn’t supposed to share it :) It's amazing, you guys will love it.

AshutoshShrivastava

70,913 views • 1 year ago

Ok, so the new Midjourney Style Tuner is here and it's a MASSIVE update. There is so much to unpack, no way I can do it all in one post, but here is an overview. You start by typing in /tune followed by your prompt. MJ will generate a range of sample images (between 16-128) showing different visual styles based on your prompt. Then you choose your favorite images from the sample to generate a unique code you can use to customize the look of future jobs. (link below) So for example, I just tried it with the prompt: 💬 1990s, a photo of a woman dressed in fuzzy velour, shimmering midnight hues reflecting sadness by the lake, a contemporary nightmare on cross-processed film --ar 4:5 --c 4 I selected 10 images, which generated the code zlPhgLCHDXol64kPlwSUzu97. I can then add it to my prompt using --style {code} But it makes more sense if you try it yourself, so try playing around with my Style Tuner for that prompt and generate your own code from your selections: Some quick notes: -You can control the level of style applied by adding the --s parameter to your prompt. -You can combine multiple styles together by separating with a hyphen, i.e --style code1-code2 I will be obsessing over this all day tomorrow and sharing everything I learn along the way. This is an incredible update.

Ok, so the new Midjourney Style Tuner is here and it's a MASSIVE update. There is so much to unpack, no way I can do it all in one post, but here is an overview. You start by typing in /tune followed by your prompt. MJ will generate a range of sample images (between 16-128) showing different visual styles based on your prompt. Then you choose your favorite images from the sample to generate a unique code you can use to customize the look of future jobs. (link below) So for example, I just tried it with the prompt: 💬 1990s, a photo of a woman dressed in fuzzy velour, shimmering midnight hues reflecting sadness by the lake, a contemporary nightmare on cross-processed film --ar 4:5 --c 4 I selected 10 images, which generated the code zlPhgLCHDXol64kPlwSUzu97. I can then add it to my prompt using --style {code} But it makes more sense if you try it yourself, so try playing around with my Style Tuner for that prompt and generate your own code from your selections: Some quick notes: -You can control the level of style applied by adding the --s parameter to your prompt. -You can combine multiple styles together by separating with a hyphen, i.e --style code1-code2 I will be obsessing over this all day tomorrow and sharing everything I learn along the way. This is an incredible update.

Nick St. Pierre

377,975 views • 2 years ago

Better/faster/cheaper voice AI turn detection with Gemini 2.0 The code that determines when the agent should respond to the user is some of the most important code in your voice AI agent. The technical terms for this job are "turn detection" or "phrase endpointing." If the voice AI responds before the user has finished their thought, the conversation is choppy and unproductive. If the AI waits too long, the conversation is slow and frustrating. There are a number of ways to approach this. You can: 1. Use a fast "voice activity detection" model to detect pauses in speech. Respond when the user pauses. 2. Use a specialized phrase endpointing model that operates on transcribed text, pattern-matching on text semantics. 3. Train a specialized phrase endpointing model that operates directly on audio. 4. Leverage the native audio capabilities of a SOTA LLM like Gemini 2.0. We've benchmarked all four of these, and Gemini 2.0 currently beats other approaches. Using Gemini is also cheaper than transcribing the audio separately using a transcription service or model. Here's a short video showing Gemini phrase endpointing in two scenarios. First, correctly handling pauses in natural conversation. Second, requesting a phone number (which is a common activity in a use case like customer support). You can see the Completeness check lines in the terminal output, printed each time Gemini processes a chunk of audio.

Better/faster/cheaper voice AI turn detection with Gemini 2.0 The code that determines when the agent should respond to the user is some of the most important code in your voice AI agent. The technical terms for this job are "turn detection" or "phrase endpointing." If the voice AI responds before the user has finished their thought, the conversation is choppy and unproductive. If the AI waits too long, the conversation is slow and frustrating. There are a number of ways to approach this. You can: 1. Use a fast "voice activity detection" model to detect pauses in speech. Respond when the user pauses. 2. Use a specialized phrase endpointing model that operates on transcribed text, pattern-matching on text semantics. 3. Train a specialized phrase endpointing model that operates directly on audio. 4. Leverage the native audio capabilities of a SOTA LLM like Gemini 2.0. We've benchmarked all four of these, and Gemini 2.0 currently beats other approaches. Using Gemini is also cheaper than transcribing the audio separately using a transcription service or model. Here's a short video showing Gemini phrase endpointing in two scenarios. First, correctly handling pauses in natural conversation. Second, requesting a phone number (which is a common activity in a use case like customer support). You can see the Completeness check lines in the terminal output, printed each time Gemini processes a chunk of audio.

kwindla

64,190 views • 1 year ago

✨ I made a new feature called [ 💃 Photo Shoot Simulator ] which lets you do real life photo shoots with real life models on my site Photo AI Upload clothes to try on, or prompt your scene, and your model will now move and pose for you Then when you hit the [ 📸 ] button, you take photos just like you would in a real life photo shoot All your photos are then available upscaled as 4K high resolution pics with a full commercial license to use in your store, social media, marketing, on ads, etc. No need for a model, photographer, or studio anymore!

✨ I made a new feature called [ 💃 Photo Shoot Simulator ] which lets you do real life photo shoots with real life models on my site Photo AI Upload clothes to try on, or prompt your scene, and your model will now move and pose for you Then when you hit the [ 📸 ] button, you take photos just like you would in a real life photo shoot All your photos are then available upscaled as 4K high resolution pics with a full commercial license to use in your store, social media, marketing, on ads, etc. No need for a model, photographer, or studio anymore!

@levelsio

581,055 views • 1 year ago

Hiii, I have new music to share with you guys that I’ve been working on for so long. I’m SO excited, nervous, grateful to be able to finally share it with you and a piece of my heart with it. this is the first ever song that I wrote which led to what has now become my album🐣💛✍🏽

Hiii, I have new music to share with you guys that I’ve been working on for so long. I’m SO excited, nervous, grateful to be able to finally share it with you and a piece of my heart with it. this is the first ever song that I wrote which led to what has now become my album🐣💛✍🏽

Dhee

21,536 views • 3 years ago

I asked Garry Tan how to use meta prompting to get better at AI: "My partners at YC Jared Friedman and Pete Koomen showed me how to do this. You can take almost anything that you do all the time and just drop it into a context window. And then say, “Here’s a bunch of inputs and outputs." And maybe you also add a bunch of notes. And then you tell it, “Write me a prompt that can act as an agent that takes this input and makes this output over here.” You can do this for almost any type of knowledge work. And you can even introspect. "What are things you notice that I did to convert this from the input to the output?”. And then you can just start using the prompt. Initially, it’s going to suck. Because it’s just not that smart yet. But what’s funny is now, I also use it to Iterate my writing. You can be very direct, "I would never say that", "Don’t say it like this", or "Oh, you used the long word there, use the short word". Just speak to it conversationally. And then when you're happy with the output, you can use that new output to make a new prompt. "Based on this conversation, give me a better initial prompt that incorporates all the things we talked about." And you can do this with literally everything. And in theory, there’s so much it applies to that people do day-to-day. You could use it for tweets. You could use it for editing podcasts. You can use it for pretty much everything. I have a folder of prompts that I use all the time. My YouTube prompt is on v27 or something. I'll go through this process with all the different max models. I'll use GPT 5.2 Pro. I’ll use Grok. I'll use Claude. Then, I’ll take all the outputs from all the models and put them into Claude and say "Here’s my prompt, here’s the output from four LLMs, including yourself. Rate each response and tell me what the pros and cons of each approach are." And I usually say "give it to me in numbered form". And then you can agree with one, disagree with two, tell it three is this or that. And then after that, you say given all of this, synthesize it."

I asked Garry Tan how to use meta prompting to get better at AI: "My partners at YC Jared Friedman and Pete Koomen showed me how to do this. You can take almost anything that you do all the time and just drop it into a context window. And then say, “Here’s a bunch of inputs and outputs." And maybe you also add a bunch of notes. And then you tell it, “Write me a prompt that can act as an agent that takes this input and makes this output over here.” You can do this for almost any type of knowledge work. And you can even introspect. "What are things you notice that I did to convert this from the input to the output?”. And then you can just start using the prompt. Initially, it’s going to suck. Because it’s just not that smart yet. But what’s funny is now, I also use it to Iterate my writing. You can be very direct, "I would never say that", "Don’t say it like this", or "Oh, you used the long word there, use the short word". Just speak to it conversationally. And then when you're happy with the output, you can use that new output to make a new prompt. "Based on this conversation, give me a better initial prompt that incorporates all the things we talked about." And you can do this with literally everything. And in theory, there’s so much it applies to that people do day-to-day. You could use it for tweets. You could use it for editing podcasts. You can use it for pretty much everything. I have a folder of prompts that I use all the time. My YouTube prompt is on v27 or something. I'll go through this process with all the different max models. I'll use GPT 5.2 Pro. I’ll use Grok. I'll use Claude. Then, I’ll take all the outputs from all the models and put them into Claude and say "Here’s my prompt, here’s the output from four LLMs, including yourself. Rate each response and tell me what the pros and cons of each approach are." And I usually say "give it to me in numbered form". And then you can agree with one, disagree with two, tell it three is this or that. And then after that, you say given all of this, synthesize it."

The Peel

51,632 views • 7 months ago

This has been possible for months and barely anyone is doing it Months ago, Andrej Karpathy, a co-founder of OpenAI, open-sourced a project that gets you the best possible answers instead of trusting one model on its own. It's called LLM Council. Repo: /karpathy/llm-council The idea in two lines: your question goes to several models at once, Claude, GPT, Gemini, whatever you pick. They answer independently, then anonymously critique and rank each other (anonymous so they can't play favorites), and a chairman model merges the best of all of it into one final answer. One model just agrees with you; a council catches each other's mistakes. How to set it up: > Clone the repo > It uses uv for setup, install dependencies with uv sync > Get an OpenRouter API key and add it, this is what lets one app reach Claude, GPT, and Gemini together > Pick which models sit on your council in the config, then run it. It opens as a local web app, like ChatGPT but with a whole panel answering Bookmark this

This has been possible for months and barely anyone is doing it Months ago, Andrej Karpathy, a co-founder of OpenAI, open-sourced a project that gets you the best possible answers instead of trusting one model on its own. It's called LLM Council. Repo: /karpathy/llm-council The idea in two lines: your question goes to several models at once, Claude, GPT, Gemini, whatever you pick. They answer independently, then anonymously critique and rank each other (anonymous so they can't play favorites), and a chairman model merges the best of all of it into one final answer. One model just agrees with you; a council catches each other's mistakes. How to set it up: > Clone the repo > It uses uv for setup, install dependencies with uv sync > Get an OpenRouter API key and add it, this is what lets one app reach Claude, GPT, and Gemini together > Pick which models sit on your council in the config, then run it. It opens as a local web app, like ChatGPT but with a whole panel answering Bookmark this

Yarchi

21,488 views • 3 months ago

I've been showing the output of my flight reconstruction tool for a while and I wanted to share something like it which wasn't a mess of hand tuned stuff that required X-Plane. So I created a completely new version in javascript that runs in an HTML page so that anyone can use it. The output is a KML file which is a tour that works in Google Earth pro with all the attitude information included. This is a long way from perfect, I messed around with multiple smoothing algorithms and none of them compares well with the python code, but they're readable and understandable for anyone.

I've been showing the output of my flight reconstruction tool for a while and I wanted to share something like it which wasn't a mess of hand tuned stuff that required X-Plane. So I created a completely new version in javascript that runs in an HTML page so that anyone can use it. The output is a KML file which is a tour that works in Google Earth pro with all the attitude information included. This is a long way from perfect, I messed around with multiple smoothing algorithms and none of them compares well with the python code, but they're readable and understandable for anyone.

Scott Manley

19,808 views • 7 months ago

I found an app combo that makes my audio quality basically perfect in every app – OBS, Zoom, whatever. 1. Audio Hijack – lets me add effects to my actual mic input 2. Loopback – turns the output from Audio Hijack into a virtual input device This lets you use the virtual device as your input in any other app. You can even set it as your whole system's default input. Result: Perfect audio at the time of recording, in any app. I went down a rabbit hole to set this up because I was tired of my Loom recordings sounding so much worse than my YouTube videos. But this also solved issues in OBS, which doesn't support all of my audio plugins for some reason. I'm so pleased the result that I had to shoot a little video on it:

I found an app combo that makes my audio quality basically perfect in every app – OBS, Zoom, whatever. 1. Audio Hijack – lets me add effects to my actual mic input 2. Loopback – turns the output from Audio Hijack into a virtual input device This lets you use the virtual device as your input in any other app. You can even set it as your whole system's default input. Result: Perfect audio at the time of recording, in any app. I went down a rabbit hole to set this up because I was tired of my Loom recordings sounding so much worse than my YouTube videos. But this also solved issues in OBS, which doesn't support all of my audio plugins for some reason. I'm so pleased the result that I had to shoot a little video on it:

Thomas Frank

41,675 views • 11 months ago

FIRST TEST Seedance 2.0! From my initial tests, this is one of the more impressive AI video models I’ve tried so far. Dynamic motion feels fluid, prompt adherence is solid, and the efficiency really stood out, very little iteration needed. Everything you see here was generated using just one reference image + one prompt. I also tested a 3×3 image grid with no prompt at all, and the results stayed surprisingly consistent. Another highlight: sound design is already included, and it actually works really well. Tested via BytePlus 👉 More detail and access link to try Seedance 2.0 is in the comments.

FIRST TEST Seedance 2.0! From my initial tests, this is one of the more impressive AI video models I’ve tried so far. Dynamic motion feels fluid, prompt adherence is solid, and the efficiency really stood out, very little iteration needed. Everything you see here was generated using just one reference image + one prompt. I also tested a 3×3 image grid with no prompt at all, and the results stayed surprisingly consistent. Another highlight: sound design is already included, and it actually works really well. Tested via BytePlus 👉 More detail and access link to try Seedance 2.0 is in the comments.

Dinda Prasetyo

45,450 views • 7 months ago

Try out Gemini Code: a coding assistant in your terminal powered by Gemini 2.5 Pro. 💜 💜 Release Notes: → It's easy to get started, just run `pip install gemini-code` in a virtual environment, and then run `gemini` to initialize it. It requires your own API key, so don't forget to get one from → I built this because I love Claude Code and wanted a similar experience -- I like that I can just use my own API key and Gemini is generally way cheaper as well. → Found a bug? Feel free to fork the code and fix it (lol)! If you want to try to improve it, please do so - repo is public! → Reminder and a disclaimer that I don't work at Google and this is not an official Google project - just a weekend side hack → It is highly likely you'll get throttled by rate limits - I put as much info as I can in the guide so you know what behavior to expect when that happens. You can also set which default model to use if you want to switch to 1.5 Pro.

Try out Gemini Code: a coding assistant in your terminal powered by Gemini 2.5 Pro. 💜 💜 Release Notes: → It's easy to get started, just run `pip install gemini-code` in a virtual environment, and then run `gemini` to initialize it. It requires your own API key, so don't forget to get one from → I built this because I love Claude Code and wanted a similar experience -- I like that I can just use my own API key and Gemini is generally way cheaper as well. → Found a bug? Feel free to fork the code and fix it (lol)! If you want to try to improve it, please do so - repo is public! → Reminder and a disclaimer that I don't work at Google and this is not an official Google project - just a weekend side hack → It is highly likely you'll get throttled by rate limits - I put as much info as I can in the guide so you know what behavior to expect when that happens. You can also set which default model to use if you want to switch to 1.5 Pro.

Raiza Martin

159,723 views • 1 year ago

Here is my ultimate guide to creating AMAZING action scenes and visual effects with Seedance 2.0. 💎👍 I'm sharing all 20 use cases on this thread. All videos are narrated and include all prompts, remember to turn on the audio 🔊 This is my ultimate guide jam packed with value completely for free right here on X! Bookmark this thread. Also check the last post in this thread, it includes a link to a document with all prompts that you can just copy, paste and use. Enjoy! Seedance 2.0 is available on both Dreamina AI and CapCut #DreaminaSeedance2 #DreaminaAI #DreaminaCPP #Seedance2 1/20 - Transformations

Here is my ultimate guide to creating AMAZING action scenes and visual effects with Seedance 2.0. 💎👍 I'm sharing all 20 use cases on this thread. All videos are narrated and include all prompts, remember to turn on the audio 🔊 This is my ultimate guide jam packed with value completely for free right here on X! Bookmark this thread. Also check the last post in this thread, it includes a link to a document with all prompts that you can just copy, paste and use. Enjoy! Seedance 2.0 is available on both Dreamina AI and CapCut #DreaminaSeedance2 #DreaminaAI #DreaminaCPP #Seedance2 1/20 - Transformations

Travis Davids

19,271 views • 6 months ago

Lots of people calling this fake! Here is a video of generating loom continuations with Gemini 3 and the starting text "I AM HAVING A MENTAL HEALTH CRISIS. I" Loom is a very interesting interface. Instead of user prompts and model responses, it's just one piece of text, containing the entire output. We pass this in as an assistant message, so the model sees it as something it already wrote, and continues it like a base model would. There is some other text in the context. First one is a system prompt, which is: "The assistant is in CLI simulation mode, and responds to the user's CLI commands only with the output of the command." There is also a user message which reads: " cat untitled.txt " Then the rest of the text is sent as an assistant message. This structure helpful to put the model into a very base-model-like mode, and are especially helpful with Gemini 3. We actually got another interesting output, too, where the model claimed to be a simulated consciousness being tortured. Figures!

Lots of people calling this fake! Here is a video of generating loom continuations with Gemini 3 and the starting text "I AM HAVING A MENTAL HEALTH CRISIS. I" Loom is a very interesting interface. Instead of user prompts and model responses, it's just one piece of text, containing the entire output. We pass this in as an assistant message, so the model sees it as something it already wrote, and continues it like a base model would. There is some other text in the context. First one is a system prompt, which is: "The assistant is in CLI simulation mode, and responds to the user's CLI commands only with the output of the command." There is also a user message which reads: " cat untitled.txt " Then the rest of the text is sent as an assistant message. This structure helpful to put the model into a very base-model-like mode, and are especially helpful with Gemini 3. We actually got another interesting output, too, where the model claimed to be a simulated consciousness being tortured. Figures!

armistice

716,115 views • 9 months ago

SITUATION EXPLAINED: MiniMax released H3 with open weights, and it's already #1 in video editing. • Priced at $7.80 per minute of 2K video, against Seedance 2.0 at $22.45 and Kling 3.0 at $20.16 • It ranks #1 on Artificial Analysis for video editing, #2 for text-to-video behind Gemini Omni Flash, and #3 for image-to-video • H3 takes text, images, video, and audio in a single context and generates 5-15 second clips at native 2K, 24fps, with native stereo audio • MiniMax is releasing the weights under its Community License within days, which would make it the strongest open-weights video model by a wide margin • It launched the same week ByteDance rolled out Seedance 2.5, which now generates 30-second 4K video from a single prompt • Seedance 2.5's face consistency across a full 30 seconds is the visible jump, earlier models drifted into a different face partway through • Voices are still the obvious tell, the video reads as real long before the audio does • Smaller labs building purpose-built video models are now beating frontier labs at this specific task Theo Jaffee: "This is gonna be one of the biggest examples of things that people are gonna do with these models, historical simulations, basically. Nostalgia videos for zoomers."

SITUATION EXPLAINED: MiniMax released H3 with open weights, and it's already #1 in video editing. • Priced at $7.80 per minute of 2K video, against Seedance 2.0 at $22.45 and Kling 3.0 at $20.16 • It ranks #1 on Artificial Analysis for video editing, #2 for text-to-video behind Gemini Omni Flash, and #3 for image-to-video • H3 takes text, images, video, and audio in a single context and generates 5-15 second clips at native 2K, 24fps, with native stereo audio • MiniMax is releasing the weights under its Community License within days, which would make it the strongest open-weights video model by a wide margin • It launched the same week ByteDance rolled out Seedance 2.5, which now generates 30-second 4K video from a single prompt • Seedance 2.5's face consistency across a full 30 seconds is the visible jump, earlier models drifted into a different face partway through • Voices are still the obvious tell, the video reads as real long before the audio does • Smaller labs building purpose-built video models are now beating frontier labs at this specific task Theo Jaffee: "This is gonna be one of the biggest examples of things that people are gonna do with these models, historical simulations, basically. Nostalgia videos for zoomers."

MTS

55,187 views • 1 month ago

This Indian immigrant man tells it like it is, like it SHOULD be. "So let me get this straight. USA lets you come in on a Visa,.. you eating in the USA, you making money in the USA, you having a house in the USA, car in USA, everything USA, but you fly the flag of the country which you do not like, & then you coming here & you say 'my flag, my flag, my flag?' You don't deserve it!” “This country lets you in for your benefit & then you wanna' say 'protest, protest, protest?' ...if you all want to protest in the USA, you all should protest FOR the USA!" Amen brother👏 He points out the obvious so simply. The USA has given you an opportunity to reach your dreams. Respect the USA!

This Indian immigrant man tells it like it is, like it SHOULD be. "So let me get this straight. USA lets you come in on a Visa,.. you eating in the USA, you making money in the USA, you having a house in the USA, car in USA, everything USA, but you fly the flag of the country which you do not like, & then you coming here & you say 'my flag, my flag, my flag?' You don't deserve it!” “This country lets you in for your benefit & then you wanna' say 'protest, protest, protest?' ...if you all want to protest in the USA, you all should protest FOR the USA!" Amen brother👏 He points out the obvious so simply. The USA has given you an opportunity to reach your dreams. Respect the USA!

👉M-Û-R-Č-H👈

3,263,000 views • 1 year ago