Gemini launches a music generation for creative expression

A New Way to Express Yourself: Gemini Now Creates Music

Since the launch of the Gemini app, we've been developing tools for creative expression through images and videos. Today, we're taking the next step: custom music generation. Today, in its beta version, the Gemini app introduces Lyria 3, the latest generative music model from Google DeepMind. Simply describe an idea or upload a photo, such as "a comedic R&B slow jam about a sock that found its pair," and Gemini will create a high-quality, memorable track in seconds. You can also ask Gemini to be inspired by something you upload.

Advantages of Lyria 3

Lyria 3 enhances audio generation in three key areas:

1. No need to provide your own lyrics: they will be created based on your request.
2. You gain more creative control over elements like style, vocals, and tempo.
3. You can create more realistic and musically complex tracks.

Here's how it works:

• Text to Track: Describe a genre, mood, or memory, and create unique tracks with lyrics or instrumental music that match your mood. “I feel nostalgic. Create a track for my mom about the wonderful times of our childhood and the memories of her fried bananas. Make it a fun Afrobeat track with a true African vibe.”
• From Photos and Videos to Track: Upload a photo or video, and Gemini will create a track with lyrics that perfectly match the mood. “Use these photos to create a track about my dog Duncan on a walk in the woods.”

The Gemini app creates 30-second tracks with custom covers designed by Nano Banana. This makes it easy to share them with friends by downloading or simply sending a link. The goal of these tracks is to offer you a unique way to express yourself.

Content creators can also experience Lyria 3 on YouTube in Dream Track. Available in the US and expanding to other countries, Lyria 3 will enhance the quality of unique soundtracks for Shorts. The ability to better tailor the soundtrack will elevate the quality of Shorts to a new level.

New Audio Verification Features

All tracks created in the Gemini app contain SynthID, our invisible watermark for identifying content created by Google AI. We are also providing more tools for identifying AI content, expanding our verification capabilities in the Gemini app to include audio alongside images and videos. Simply upload a file and ask if it was created using Google AI, and Gemini will check for SynthID and provide an answer.

Responsible Development of Generative AI

Since the launch of Lyria in 2023, we have been committed to developing this technology responsibly, collaborating with the music community. We've learned a lot through these collaborations and experiments, such as the Music AI Sandbox, and we closely monitor copyright compliance and partner agreements when training Lyria 3.

Music generation with Lyria 3 is intended for original self-expression, not for imitating existing artists. If your request mentions a specific artist, Gemini interprets this as broad creative inspiration and will create a track that matches the style or mood. We have also implemented filters to check results against existing content. We understand that our approach may be imperfect, so you have the option to report content that may infringe on your rights or the rights of others. Additionally, users must adhere to our Terms of Use and Prohibited Use Policy for Gen AI, which prohibit violations of intellectual property and privacy rights of others.

Lyria 3 is available in the Gemini app for all users aged 18 and over in English, German, Spanish, French, Hindi, Japanese, Korean, and Portuguese, with plans to expand quality and coverage to other languages, starting with the desktop version today and the mobile app in the coming days. Google AI Plus, Pro, and Ultra subscription users will receive higher limits.

Our goal with music generation in the Gemini app is to help you add a fun, unique soundtrack to your everyday life. Try it today at gemini.google.com.