Google DeepMind's AI for creating full-length, high-fidelity songs from text or images.
Lyria 3 is our most advanced music generation model yet. It helps you play with new ideas and compose whatever you want to hear next.

Teams use Lyria to generate cohesive tracks matching their project's exact duration.
Teams use Lyria to compose with images by transforming them into custom high-fidelity tracks.
Teams use Lyria to define realistic vocal styles and acoustic preferences.
Teams use Lyria to generate vocals in different languages and create music across genres.
Teams use Lyria to write their own lyrics or generate them around specific subjects or themes.
Teams use Lyria to export professional-grade audio for their projects.
Create your own music from lyrics, text, or images with official InsMelo's free AI song generator & maker online. Transform ideas into royalty-free songs in 400+ genres. Easy-to-use for creators, marketers, and musicians.
KikiVoice is a free, browser-based AI voice cloning platform — no sign-up, 75+ languages, and up to 99% voice similarity, with three built-in models (Core, Pro, Multilingual).
Create original songs from text prompts or lyrics with MuseGen's AI music generator, then preview your track online.
Extract vocal and music