Google debuted its three newest generative AI models for media creation, Imagen3, Music AI Sandbox, and Veo, during Google I/O 2024.
The tech giant says that the three newest AI tools are its best yet, offering even higher quality offerings compared to previous models. This is especially true for its new image creation tool, which Google claims can now understand the prompts that humans write even more clearly.
We’re introducing Imagen 3: our highest quality text-to-image generation model yet.
It produces visuals with incredible detail, realistic lighting and fewer distracting artifacts.
From quick sketches to very high-res imagery, here’s a look at what it can create. #GoogleIO pic.twitter.com/XMrQYGeSiO
— Google DeepMind (@GoogleDeepMind) May 14, 2024
This allows Imagen3 to create images that take everything about the prompt into account, including the smallest details. The generative AI model can also now create text more reliably than previous models like Imagen2, which launched in Google’s chatbot back when it was still called Bard, and Google even claims that it works better than other image generation models out there.
On the audio side, Google has been working with real life producers and musicians like Marc Rebillet to create a system that works for the artist, giving them the tools they need to succeed right in the Music AI Sandbox.
Tech. Entertainment. Science. Your inbox.
Sign up for the most interesting tech & entertainment news out there.
By signing up, I agree to the Terms of Use and have reviewed the Privacy Notice.
Together with @YouTube, we’ve been building Music AI Sandbox, a suite of AI tools to transform how music can be created.
To help us design and test them, we’ve been working closely with musicians, songwriters and producers. ↓ #GoogleIO pic.twitter.com/pMLa3aCveu
— Google DeepMind (@GoogleDeepMind) May 14, 2024
According to the video, you can simply throw in a prompt, add in your own music if you want, and see what the AI does to customize it.
But Music AI Sandbox and Imagen3 aren’t the only generative AI that Google has been working on for media creation. The tech giant has also created a new video generation model called Veo, which Google says can really focus in on the smallest details, providing a great tool for video artists to utilize.
Introducing Veo, our new generative video model from @GoogleDeepMind.
With just a text, image or video prompt, you can create and edit HQ videos over 60 seconds in different visual styles. Join the waitlist in Labs to try it out in our new experimental tool, VideoFX #GoogleIO pic.twitter.com/RnMsWu9s1q
— Google (@Google) May 14, 2024
Google showed off some of the capabilities that Veo has, showing how it can build cinematic shots with just a few keywords in your prompt. The video quality is honestly astounding based on what we’ve seen so far, and it could open a lot of new doors for creators.
Google also showcased its plans to bring more Gemini tools to Android 15, and plans to allow for the creation of GPT-like “Gems,” which you can use to customize how you use Gemini going forward.
All of these new tools will be available to developers as part of Google Labs, and the waitlist is now open for those who want to sign up. Google hasn’t shared any wider plans to release of the new tools just yet. Hopefully, we’ll learn more about their individual capabilities in the coming months.
