Google continues to innovate with the introduction of new AI tools.
Recently, on Wednesday, the tech giant unveiled a significant upgrade to its AI music generator, known as Lyria 3. This latest version makes music generation much more accessible, allowing users to produce high-quality, 30-second audio tracks across various genres. From catchy jingles and relaxing lo-fi beats to intricate arrangements and full songs, Lyria 3 expands creative possibilities.
This Tweet is currently unavailable. It might be loading or has been removed.
How to Try Lyria 3 for Free
Lyria 3 is currently available in the Gemini desktop app, with a mobile app rollout planned for the near future. It is free to all users aged 18 and above. The tool is accessible in multiple languages, including English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese, with more languages anticipated in the future.
To get started, open the Gemini app. Navigate to the “Tools” dropdown menu and select the “Music” option, marked by a music note icon. Google has also provided a helpful Lyria 3 prompt guide to assist users in crafting their prompts.
How Lyria 3 Works
Google showcases an example of music generated by Lyria 3.
Credit: Google
Lyria 3 allows users to generate a 30-second music track based on a simple text, image, audio, or video prompt. Users can specify the music style, vocals, and tempo directly within their prompts. There is no need to provide lyrics, as Lyria 3 can generate them automatically. Additionally, the Nano Banana feature can even create album artwork to accompany the music.
For those looking to create music that matches a specific visual element, users can upload a relevant photo or video alongside their prompts. For instance, content creators can quickly produce a custom music track to complement their videos.
Mashable Light Speed
There’s also a template option featuring dynamic suggestions for those who need creative inspiration while crafting their AI-generated song.
According to Google, all tracks produced within the Gemini app include a SynthID watermark, distinguishing them as AI-generated content.
While free users can access Lyria 3, subscribers of Google AI Plus and AI Ultra will benefit from higher usage limits.
Addressing Copyright Concerns
The AI music industry has faced scrutiny from musicians and music publishers regarding the training of models like Lyria 3 on copyrighted material without appropriate permission or compensation. Several lawsuits are currently in progress to address these issues, with many in the industry arguing that training models on copyrighted works falls under fair use.
Although large tech companies often refrain from disclosing specific training data used for their advanced models, Google has stated that it remains “mindful” of copyright issues. A blog post announcing Lyria 3 mentioned that the company is conscious of copyright and partnership agreements during its training process.
Furthermore, Google emphasizes that Lyria is not intended to replicate existing artists. Musicians retain the right to report content if they feel their rights have been violated.
Google’s blog post clarifies:
Music generation with Lyria 3 aims for original expression, not imitation of existing artists. If a prompt references a specific artist, Gemini will interpret this as general creative inspiration and produce a track in a similar style or mood. Additionally, we have mechanisms in place to verify outputs against existing content. While we acknowledge that our approach may not be foolproof, users can report any content they believe violates their rights, as users must comply with our Terms of Service and Generative AI prohibited use policies, which prohibit violations of intellectual property and privacy rights.
Disclosure: Ziff Davis, Mashable’s parent company, filed a lawsuit against OpenAI in April 2025, alleging copyright infringement in the development and operation of its AI systems.
