Google is expanding the creative capabilities of its Gemini app with the introduction of Lyria 3, a new AI-powered music generation tool. Rolling out to users today, Lyria 3 allows anyone to create 30-second musical tracks from simple text prompts, image uploads, or even existing video clips.
Developed by Google DeepMind, Lyria 3 represents a significant leap forward in AI music creation. Unlike previous iterations, the model now automatically generates lyrics alongside the instrumental track, eliminating the need for users to write their own. The system offers control over style, vocals, and tempo, aiming to produce “more realistic and musically complex tracks,” according to Google.
The launch comes amidst growing interest – and some concern – surrounding AI’s role in the creative arts. Google is proactively addressing potential copyright issues, emphasizing that Lyria 3 is “designed for original expression, not for mimicking existing artists.” The company states it has been “very mindful of copyright and partner agreements” during the model’s training process. If a user specifically requests a track in the style of a particular artist, Gemini will interpret that as broad creative inspiration, rather than attempting a direct imitation. Google has implemented filters to scan outputs against existing copyrighted content.
“We recognize that our approach might not be foolproof,” Google acknowledges, providing a reporting mechanism for users to flag content that potentially infringes on existing rights. This feature allows users to submit audio files to Gemini to determine if they were generated using Google AI, joining similar verification tools for images and videos.
The process is designed to be intuitive. Users can simply open the Gemini app and select “Music” from the “Tools” menu. Prompts can range from specific genre requests – “an afrobeat track for my mother about the great times we had as kids and the memories of her home-cooked plantains” – to more abstract concepts. Uploading a photograph or video allows Gemini to analyze the visual content and generate a track with lyrics that complement the mood and imagery.
The resulting 30-second tracks are accompanied by custom cover art generated by Nano Banana, another Google AI model. Users can then easily share their creations by downloading the audio file or obtaining a shareable link.
Currently, Lyria 3 is available to all Gemini app users aged 18 and over in English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese. Subscribers to Google AI Plus, Pro, and Ultra plans will receive higher usage limits. Google plans to expand language support and improve the overall quality of the generated music in future updates.
The arrival of Lyria 3 positions Google as a key player in the rapidly evolving landscape of AI-driven music creation. While the technology is still in its early stages, it offers a glimpse into a future where anyone can easily generate personalized soundtracks for their lives, regardless of musical training or experience. The emphasis on originality and copyright protection suggests Google is attempting to navigate the ethical and legal complexities of this new frontier responsibly, though the long-term impact on the music industry remains to be seen.
