Google's Gemini assistant can now compose original music tracks from text descriptions or personal photos, marking a expansion of its capabilities beyond text-based assistance. The new Lyria 3 feature transforms simple prompts into 30-second soundtracks complete with custom lyrics and cover art, available immediately on desktop with mobile support arriving within days.
Users aged 18 and older can generate tracks by describing genres, moods, or uploading images for visual inspiration. The system automatically crafts verses based on provided context without requiring user-written lyrics, offering control over vocal tones, tempos, and specific styles.
Each generated track includes custom artwork created by Google's Nano Banana image model, making them shareable via download links.
The rollout positions Gemini as a direct competitor to Apple's newly announced Playlist Playground feature in Apple Music, which creates playlists from text prompts using Apple Intelligence. Both companies are targeting Spotify's territory as the streaming service faces pressure to develop its own AI mixing capabilities following Google's announcement.
Google has implemented several safeguards to address music industry concerns about copyright infringement. The system prohibits lifting content from specific artists and treats musician names as "broad creative inspiration" rather than direct imitation.
Every generated track contains SynthID watermarks developed by Google DeepMind for identification as AI-generated content.
Free users can create up to 10 tracks daily, while paid subscribers on Google AI Plus, Pro, and Ultra tiers receive between 20 and 100 daily generations depending on their plan. The feature launched alongside the Pixel 10a announcement this week and is currently available in English, German, Spanish, French, Hindi, Japanese, Korean and Portuguese.
Lyria 3 also powers YouTube's Dream Track feature for Shorts creators, which was initially available only in the United States but is now rolling out to creators in other countries. This integration provides professional-grade tools for short-form video content creation alongside the standalone Gemini app functionality.
The music generation capability represents Google's latest move to demonstrate revenue potential from AI investments following widespread praise for its Gemini 3 model released last November. That earlier release reportedly prompted OpenAI CEO Sam Altman to declare a "code red" response to accelerate ChatGPT improvements.
