Google continues to release new AI tools.
On Wednesday, Google launched a new AI music generator called Lyria 3. A huge improvement over previous versions of the model, as it makes music production much easier for users. Lyria 3 can create reliable, 30-second audio tracks in a variety of genres, from short beats and lo-fi beats to complex arrangements and songs.
This Tweet is currently unavailable. Either it is loading or it has already been downloaded.
How to try Lyria 3 for free
Google says Lyria 3 is available now on the Gemini desktop app (it will be released on the mobile app in the coming days), and it’s free for all users 18 and older. Lyria can be used in English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese, with support for other languages to come.
To get started, just go to the Gemini app. In the “Tools” drop-down menu, look for the “Music” option, which is accompanied by a music note icon. Google has also released a quick guide for Lyria 3 to help users get started.
How Lyria 3 works
Google has provided this example for Lyria 3 music generation.
Credit: Google
Lyria 3 can create a short 30-second song based on simple text, image, audio, or video information. Users can control the music style, vocals, and tempo all within the prompt. In addition, users no longer have to provide their own songs. Lyria 3 can create them based on information. Nano Banana will produce the accompanying album artwork.
If users want to create music to accompany a particular photo or video, they can upload it along with their information. For example, social creators can upload a video and create a custom music track to accompany it in seconds.
Mashable Light Speed
In addition, there is a template option with flexible suggestions for people who would like to create an AI song but need some inspiration.
According to Google, all tracks produced within the Gemini app have an embedded SynthID watermark, marking the content as AI-generated.
While free users will have access to Lyria 3, paying Google AI Plus and AI Ultra subscribers will get higher usage limits.
What about copyright?
Many artists and music publishers have criticized the AI industry for training models like Lyria 3 on copyrighted works without permission or payment, and many lawsuits are working their way through the court system based on these concerns. Many AI companies argue that training models in patented applications are protected by fair use.
While Big Tech companies typically refuse to comment on the specific training materials used for their frontier models, Google said it is “aware” of music copyrights. Google’s blog post announcing the new model said the company “was very careful about copyright agreements and partner agreements as we developed Lyria 3.
In addition, Google says Lyria is not designed to imitate existing artists, and artists can report content if they believe it infringes their rights.
The blog post says:
Music generation with Lyria 3 is designed for original expression, not imitation of existing artists. If your input mentions a certain artist, Gemini will take this as a broad creative inspiration and create a track that shares the same style or mood. We also have filters available to test the output against existing content. We recognize that our practices may not be reasonable, so you can report content that may violate your rights or the rights of others. Additionally, in order to use our products, users must comply with our Terms of Service and Gen AI’s prohibited use policies, which prevent infringement of others’ intellectual property rights and privacy rights.
Disclosure: Ziff Davis, Mashable’s parent company, in April 2025 filed a lawsuit against OpenAI, alleging that it infringes Ziff Davis’s copyright in training and using its AI programs.
