Gemini’s Lyria 3 Now Lets Everyone Create Music
▼ Summary
– Google DeepMind has launched Lyria 3, its most advanced music generation model, which automatically creates lyrics and offers control over style, vocals, and tempo.
– The model is designed for original expression, not mimicking artists, and uses filters to avoid copyright infringement by treating artist names as broad inspiration.
– Users can generate 30-second tracks in the Gemini app by describing a genre, mood, or memory, or by uploading a photo or video for lyrical inspiration.
– All generated audio includes a SynthID watermark for verification, allowing users to check if a file was created with Google AI.
– Lyria 3 is currently available for users 18+ in several languages, with higher limits for subscribers, and Google plans future expansions in language support and quality.
Google’s Gemini app has introduced a powerful new feature for music creation, powered by the advanced Lyria 3 model. This tool allows users to generate original 30-second music tracks complete with AI-generated lyrics and custom cover art, directly from a simple text prompt. The model is designed to foster original expression rather than mimic existing artists, offering users control over elements like style, vocals, and tempo to produce musically complex results.
Accessing the feature is straightforward. Within the Gemini app, users navigate to the “Tools” menu and select “Music.” From there, they can type a description of the music they want to hear. Prompts can be highly specific, requesting a particular genre, mood, or even referencing a personal memory or inside joke. For instance, someone could ask for “a fun afrobeat track with a true African vibe” about nostalgic childhood memories. Alternatively, users can upload a photo or video clip, and Gemini will compose a track with lyrics that match the visual content’s mood.
A key aspect of Lyria 3 is its approach to copyright and artistic integrity. The system is built to interpret mentions of specific artists as broad creative inspiration, generating music in a similar style or mood rather than producing imitations. Google has implemented filters to check outputs against existing content and embeds all generated tracks with a SynthID watermark for verification. Users can upload any audio file to Gemini to ask if it was created with Google’s AI, providing a layer of transparency. The company also encourages reporting of any content that may infringe on rights, acknowledging that no system is entirely foolproof.
Currently, the Lyria 3 music feature is available to all users aged 18 and older in several languages, including English, Spanish, French, German, Hindi, Japanese, Korean, and Portuguese. Subscribers to Google’s AI Plus, Pro, and Ultra plans will benefit from higher usage limits. Looking ahead, Google plans to expand the tool’s availability to more languages and continue improving the overall quality of the music it generates.
(Source: NewsAPI Tech Headlines)
