Google's Gemini: Crafting Music from Text and Images
product#generative ai📝 Blog|Analyzed: Feb 19, 2026 07:30•
Published: Feb 19, 2026 07:20
•1 min read
•ASCIIAnalysis
Google is expanding its Gemini app with a beta music generation feature, powered by the Lyria 3 model. This innovative tool allows users to create 30-second songs from text prompts or visual inputs like photos and videos, opening exciting new avenues for creative expression. The integration of SynthID for AI-generated content verification further emphasizes Google's commitment to responsible innovation.
Key Takeaways
- •Gemini's new feature generates 30-second songs from text and images.
- •The system uses Google DeepMind's Lyria 3 model, enhancing sound quality and complexity.
- •AI-generated music includes a SynthID watermark for identification and verification.
Reference / Citation
View Original"Users can input an image such as 'a comical R&B song that plays when you meet your soulmate', and in a matter of seconds generate a song with lyrics or instrumental music."