Build Music Generation into Your Apps with Lyria 3 Models on Vertex AI
Google has made its Lyria 3 family of music generation models available on Vertex AI in public preview, allowing developers to generate high-quality and high-fidelity stereo audio from text prompts and images.
Why it matters
Lyria 3 models enable developers to build studio-quality audio production directly into their apps, unlocking new creative possibilities.
Key Points
- 1Lyria 3 Pro generates complete compositions up to 3 minutes long with specific structural elements
- 2Lyria 3 generates tracks up to 30 seconds long for rapid prototyping, social media assets, and short-form audio
- 3Supports multi-modal input (text prompts and reference images), vocals and lyrics, and flexible compositions
Details
The Lyria 3 models on Vertex AI offer two distinct options - Lyria 3 Pro for generating complete compositions up to 3 minutes long with specific structural elements like intros, verses, choruses, and bridges, and Lyria 3 for generating shorter 30-second tracks ideal for rapid prototyping and social media. These models can generate high-quality, high-fidelity stereo audio from text prompts or by using reference images to guide the story, mood, and style. They also support the generation of vocals and timed lyrics, or can create pure instrumental soundbeds. The flexible composition capabilities allow developers to create full songs with distinct sections. Lyria 3 models are available through the Vertex AI API and Vertex AI Media Studio.
No comments yet
Be the first to comment