Google’s AI Music Creator Launches in the Gemini App

Gemini has been upgraded with a new feature that allows users to create AI-generated music, thanks to DeepMind’s latest audio model. The beta version of Lyria 3 is now available in the Gemini app, enabling users to craft 30-second tracks based on text, images, and videos, all within the chatbot interface.

Starting today, this music creation tool is accessible globally in multiple languages, including English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese, with plans for further expansion. However, access is restricted to users aged 18 and older.

With Lyria 3’s text-to-music functionality, users can generate songs by describing genres, moods, or personal memories. For instance, you could request an “Afrobeat track for my mother about the great times we had growing up.” The music generator can produce both instrumental pieces and songs with lyrics that are automatically created based on user input. Additionally, users can upload photos and videos, which Gemini uses to generate tracks with lyrics that match the intended vibe.

The aim of these tracks is not to create a musical masterpiece but to provide a fun and unique way for users to express themselves. Gemini will also include custom cover art generated by Nano Banana for the songs created in the app, making them easier to share and download. Furthermore, Lyria 3 will be integrated into YouTube’s Dream Track tool, allowing creators to produce custom AI soundtracks for Shorts.

Initially, Dream Track and Lyria showcased the ability to replicate the style and voice of well-known artists. Google has emphasized that it has been “very mindful” of copyright issues during the development of Lyria 3, ensuring that the tool is intended for original expression rather than imitating existing artists. When users request a specific artist, Gemini will generate a track that captures a similar style or mood while employing filters to ensure outputs do not infringe on existing content.

Lyria has been available since 2023, but its access was previously limited to Google Cloud’s Vertex platform. This new rollout to the Gemini app enhances accessibility, although it comes after similar offerings from platforms like TikTok and Microsoft’s Copilot, which have already introduced their own AI music creation tools.