Google’s Gemini App Now Lets Users Generate AI Songs With DeepMind’s Lyria 3

Google is expanding the creative powers of its Gemini app in a major way. The company announced this week that it is introducing Gemini AI music generation, allowing users to create original songs directly through simple AI prompts.

The new feature is powered by DeepMind’s Lyria 3 music-generation model, which Google says produces more realistic and complex tracks than previous versions. The tool is currently available in beta, but it signals Google’s growing ambition to make Gemini a complete creative assistant, not just a chatbot.

With this update, Gemini users can generate short songs, complete with lyrics, cover art, and even mood-based soundtracks for uploaded media.

How Gemini’s AI Music Feature Works

Using the new music-generation option is simple: users describe the type of song they want, and Gemini generates a track in response.

For example, Google suggested a playful prompt like:

Create a comical R&B slow jam about a sock finding its match.

Gemini will then produce a 30-second music clip, complete with lyrics and AI-generated cover art created using Google’s Nano Banana system.

This makes Gemini one of the few mainstream AI apps offering an all-in-one package: music + lyrics + visuals in a single workflow.

Songs That Match Photos and Videos

One of the most interesting additions is Gemini’s ability to generate music based on uploaded content.

Users can upload a photo or video, and Gemini will create a song that matches the mood of the media file.

That means a sunset clip could result in a calm ambient track, while an energetic travel video might generate upbeat pop music.

This feature could appeal strongly to:

• Social media creators
• Short-form video editors
• Casual users experimenting with sound
• Marketing teams building quick creative assets

Google says Lyria 3 improves not only music quality but also user control. With the updated model, users can adjust creative elements such as:

• Musical style
• Vocal presence
• Tempo
• Mood and genre direction

This added flexibility makes Gemini’s tool feel less like a gimmick and more like an early-stage music studio assistant.

Lyria 3 Comes to YouTube Through Dream Track

Alongside Gemini, Google is also expanding Lyria 3 access for YouTube creators via Dream Track, a feature that helps creators generate AI music for videos.

Previously, Dream Track was only available to creators in the United States. With this release, Google is now rolling it out globally, opening access to a much larger creator community.

This could significantly change how background music is produced for content across YouTube.

Can Users Copy Artists? Google Says No, But It’s Complicated

One of the biggest controversies around AI music is whether users can imitate famous artists. Google claims the system is designed for original creation, not direct replication. In its blog post, the company explained:

Music generation with Lyria 3 is designed for original expression, not for mimicking existing artists.

However, Google confirmed that if users include an artist’s name in a prompt, Gemini may generate music with a similar mood or style, though not an outright clone.

This raises ongoing concerns about how easily AI can blur the line between inspiration and imitation.

SynthID Watermarking to Identify AI Music

To address authenticity and copyright concerns, Google is embedding a SynthID watermark into every track generated by Lyria 3.

This watermark helps identify music as AI-generated, even if it is shared outside Gemini or YouTube.

Google is also adding detection features within Gemini itself:

• Users will be able to upload a track
• Ask Gemini whether it was AI-generated
• Receive confirmation through SynthID analysis

This is part of Google’s broader push toward transparency in generative AI.

The music-generation feature is rolling out to all Gemini users aged 18 and above worldwide.

It supports multiple languages at launch, including English, German, Spanish, French, Hindi, Japanese, Korean and Portuguese. This makes it one of the most internationally accessible AI music tools so far.

Why AI Music Still Divides the Industry

AI-generated music is quickly becoming one of the most debated areas of generative technology. On one side, platforms like YouTube and Spotify are exploring monetization models and partnerships with music labels.

On the other, AI companies are facing lawsuits over copyrighted training material, unauthorised use of artist catalogs and revenue distribution.

Streaming services like Deezer have already launched detection tools to flag AI-generated tracks and reduce fraudulent streaming activity. Google’s SynthID approach may help, but the legal and ethical debates are far from settled.

What This Means for the Future of Gemini

With music generation now joining text, images, and video understanding, Gemini is evolving into a broader creative platform. The next big question will be whether tools like Lyria 3 empower creativity responsibly, or accelerate conflicts over ownership and originality.