Gemini has launched a new feature using Lyria 3, Google DeepMind's generative music model, to create 30-second songs with lyrics from text prompts or photos. This innovation aims to make music creation more accessible and playful, rather than replacing musicians.
Features and Functionality
Lyria 3 allows users to generate lyrics automatically and offers enhanced control over style, vocals, and tempo, resulting in more realistic, layered tracks. Users can input a prompt or upload an image or video to receive a polished mini-track, complete with custom cover art by Nano Banana.
Integration with YouTube Shorts
The model will also power Dream Track on YouTube Shorts, providing customizable backing tracks and lyrics for short videos. Google has ensured the model respects copyright and partner agreements, incorporating filters to check for existing content and using SynthID, an invisible watermark, for audio identification and verification.
Availability and Access
Lyria 3 is available in the Gemini app for users aged 18 and older, supporting multiple languages. It launches on desktop first, with mobile access coming soon. Google AI Plus, Pro, and Ultra subscribers benefit from higher usage limits.
