Lyria 3 Pro: Extended Music Generation with Structural Awareness
Key Points
- 3-minute track generation with structural composition awareness
- Integration across multiple Google products and developer platforms
- Built-in SynthID watermarking and content protection measures
Summary
Google DeepMind has released Lyria 3 Pro, an advanced music generation model that creates tracks up to 3 minutes long with enhanced structural awareness. The model understands musical composition elements like intros, verses, choruses, and bridges, enabling more sophisticated music creation with complex transitions.
Key Points
- Extended Track Length: Lyria 3 Pro generates music up to 3 minutes long, significantly longer than previous versions
- Structural Awareness: Model understands musical composition elements and can be prompted for specific song sections
- Multi-Platform Integration: Available across Vertex AI, Google AI Studio, Gemini API, Google Vids, Gemini app, and ProducerAI
- Enterprise Scale: Public preview on Vertex AI enables businesses to generate high-fidelity audio at scale
- Developer Access: Available in Google AI Studio and Gemini API for building creative applications
- Content Protection: All outputs include SynthID watermarking and employ filters to prevent mimicking existing artists
- Industry Collaboration: Developed with input from Music AI Sandbox partners and used by Grammy-winning producers