Google Adds AI Music Generation to Gemini App Using Lyria 3: Here’s How to Use

Key Highlights:

  • All AI-generated songs carry a SynthID watermark for identification.
  • Google adds AI-powered music generation to the Gemini app using DeepMind’s Lyria 3 model.
  • Users can create 30-second songs with lyrics, cover art, and mood-based audio.
  • Dream Track expands globally for YouTube creators beyond the U.S..

Google has added music-generation capabilities to the Gemini app. The feature uses Lyria 3, a new music-generation model developed by DeepMind. The rollout is global and currently in beta.

The update allows users to generate short music tracks by simply describing what they want. Gemini creates a 30-second song, complete with lyrics and AI-generated cover art.

How does music generation work in the Gemini app?

Users can type a prompt describing the song they want. The app then generates music based on style, tempo, and mood. Prompts can be playful, cinematic, or emotional.

Gemini also supports media-based generation. Users can upload a photo or video, and the app will generate a song that matches the mood of the visual content. This makes the tool useful for creators, students, and casual users exploring audio storytelling.

Google says Lyria 3 produces more realistic and complex compositions compared to earlier models. Users can tweak elements like vocals, tempo, and overall style.

Can Gemini copy real artists or music styles?

Google says no direct imitation is allowed. Users cannot fully mimic a specific artist. If an artist’s name appears in a prompt, Gemini treats it as broad inspiration rather than replication.

According to Google, filters are in place to prevent outputs that closely resemble existing songs. The company says the goal is original expression, not copying known artists or tracks.

What is Dream Track and why does this matter?

Alongside Gemini, Google is expanding Lyria 3 to YouTube through Dream Track. The tool helps creators generate AI music for videos.

Dream Track was previously limited to U.S. creators. With this update, it is now available globally through YouTube. This expands access to AI-generated background music for creators worldwide.

How is Google handling AI music identification?

All songs generated with Lyria 3 will include a SynthID watermark. This watermark identifies AI-generated audio without affecting sound quality.

Google is also adding detection tools inside Gemini. Users can upload a music file and ask whether it was created using AI.

This move comes as platforms like Deezer and Spotify face growing concerns around AI music, copyright disputes, and fraudulent streams.

Availability and supported languages

Music generation is rolling out to all Gemini users aged 18 and above. The feature supports English, Hindi, German, Spanish, French, Japanese, Korean, and Portuguese.

As Gemini continues to expand beyond text and images, music generation marks another step in making the app a multi-modal AI tool.

73 Views