tech
February 18, 2026
Record scratch—Google's Lyria 3 AI music model is coming to Gemini today
With a simple prompt, you can generate 30 seconds of something like music.
TL;DR
- Google's Lyria 3 AI music generation model is now available in the Gemini app and web interface.
- Users can create 30-second music tracks by describing their desired genre, mood, and theme, or by uploading an image.
- The AI can generate lyrics for the tracks, and each creation includes an album cover-style image from the Nano Banana model.
- Lyria 3 tools are also being integrated into Google's Dream Track toolkit for YouTube Shorts.
- Google has embedded an audio version of its SynthID watermarking technology into Lyria 3-generated tracks.
- The model is designed to respect copyright, taking artist names as broad creative inspiration rather than direct imitation, though users can report content that imitates too closely.
- Lyria 3 supports multiple languages including English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese, with plans for more.
- While all users have some access, AI Pro and AI Ultra subscribers will have higher usage limits.
Continue reading
the original article