Google DeepMind’s Gemini AI Transforms Music Generation With Lyria 3

In Focus

• The new tool creates music tracks using text prompts
• Lyria 3 can generate music tracks from videos and photos
• The AI music generation tool improves on Lyria 3

Google DeepMind has launched an AI-powered music generation model called Lyria 3. According to Gadgets360, the Gemini AI music generation model allows users to create high-quality music tracks using text prompts. The AI model is designed to integrate into the Gemini app on both iOS and Android devices.

Google Targets Creators With New Tool

DeepMind is positioning the new Google AI music feature as a creator-focused tool that simplifies content production for digital storytelling, videos, and social media. According to the tech company, content creators can use Lyria 3 to generate original tracks, up to 30 seconds long.

To generate music, users should describe their desired mood, genre, instruments, or tempo.
The AI tool takes the text prompts and translates them into audio clips with a polished sound output and layered instrumentation.

“The Gemini app creates 30-second tracks with custom cover art generated by Nano Banana. This makes it easy to quickly share with friends by downloading or simply clicking the share link. The goal of these tracks isn’t to create a musical masterpiece, but rather to give you a fun, unique way to express yourself,” Senior Product Manager at Gemini app Joël Yawili stated in a blog post.

Besides turning text prompts into music tracks, creators can use Google’s Gemini AI music generation feature to generate music tracks from videos and photos.

For instance, a user can take a short travel clip and use it as a creative input on the tool to generate a soundtrack that matches its atmosphere and visual tone. Users can add customized AI-generator cover art to every generated track, enabling creators to package and share their content with ease.

Lyria 3 Improves on the Lyria Model

Google’s Lyria 3 AI music model is a significant improvement of Lyria which launched in 2023.
According to the tech company, the new model improves audio generation by eliminating the need for users to provide their own lyrics.

The model also offers users more control over creative elements like vocals, style, and tempo, and enables them to generate more realistic and complex tracks. Google says creators can explore the new AI music generation tool on YouTube’s Dream Track, which is available in the U.S. On the platform, the tech company says Lyria 3 can enhance the quality of soundtracks used in each unique Short.

“Whether it’s creating a lyrical verse or a vibey backing track, being able to better customize the soundtrack will take creators’ Shorts to the next level,” Yawili added.

Lyria 3 Comes With Verification Capabilities

Google has integrated its SynthID watermarking system directly into Lyria 3’s audio creations to strengthen trust in AI-produced media. This technology discreetly embeds identifying signals within the sound file itself, allowing the content to be recognized as AI-generated.

The watermark cannot be noticed by listeners. However, it can be detected through verification tools. Lyria 3 is currently accessible through the Gemini app for users over 18 years of age in multiple languages, including English, German, French and Spanish.