Introducing Lyria 3: Our new music model

By Google DeepMind

Categories: AI, Product

Summary

DeepMind's new Lyria 3 music model lets you turn images into high-fidelity audio, customize genres, dynamics, and vocals across multiple languages - a powerful creative tool for brands, musicians, and audio creators.

Key Takeaways

1. Create high-fidelity audio from images to express your brand through personalized music.
2. Blend and customize music genres, dynamics, tempo, and realistic vocals in multiple languages.
3. Export crisp, clear audio that is watermarked and identified as an AI creation.
4. Leverage Lyria 3 as a 'musical collaborator' to compose any type of audio you desire.
5. Experiment with new musical ideas and turn detailed prompts into tracks with natural flow.
6. Remember and recreate the sounds of places you love with a tune you won't forget.

Topics

• AI-Generated Music
• Multimodal AI Models
• Brand Audio Identity
• Music Customization
• Audio Creation Toolkits

Transcript Excerpt

You’re listening to Lyria 3, from Google Our most advanced music generation model yet Play with new ideas, turn detailed prompts into tracks with natural flow Transform your images into high-fidelity audio to express your brand through music, or remember a place you love, with a tune you won’t forget And you can direct the details Pick your favourite genre, or blend new ones, get specific with dynamics and tempo, and define realistic vocals, in a range of languages When you’ve found your sound, ...