Google has added a major new feature to its Gemini app. Starting with the April 2026 update, users can now create original music directly inside Gemini. The feature uses Google DeepMind’s latest Lyria 3 model, which was trained on over 200 million songs. For competitors like Suno and Udio, this is a serious challenge. Google is not just adding a music tool. It is bringing a complete creative platform to 750 million users.
How It Works
Using the new feature is simple. Open the Gemini app and type a description of the song you want. You can also upload a photo or video. Gemini will analyze the image and create music that matches the mood. Within seconds, you get a 30-second track with vocals, lyrics, and instruments.
For example, you can say “create a happy pop song about summer” or upload a sunset photo and ask for matching music. Gemini handles the rest. The AI writes the lyrics, composes the melody, and generates the vocals. You do not need any musical skills.
The feature is powered by Lyria 3, Google DeepMind’s newest music generation model. Earlier versions of Lyria were already used in YouTube’s Dream Track feature. Lyria 3 brings major improvements including automatic lyrics, better vocal quality, and support for images and videos as input.
What Makes Lyria 3 Different
Google is positioning this feature as a new way to express creativity. Unlike other AI music tools that focus on instrumental tracks, Lyria 3 generates complete songs with lyrics and vocals from the start.
Here are the key improvements in Lyria 3:
- Automatic lyrics: The model writes original lyrics based on your prompt. You do not need to write anything yourself.
- Vocal generation: Lyria 3 creates natural-sounding vocals with proper emotion and articulation.
- Image and video input: Upload a photo or video and the AI will compose music that matches the visual mood.
- High-quality audio: The output is 48kHz stereo audio with professional-grade clarity.
- Cover art: Each track gets automatic album art generated by Google’s Nano Banana image model.
According to Digital Music News, Lyria 3 was trained on about 200 million songs. This is roughly 50 times more than the previous Lyria 2 model. The audio quality has also improved from 16-bit to 24-bit depth, matching the standards used by YouTube Music and other major platforms.
Availability and Access
The music feature is rolling out in beta to Gemini users worldwide. You must be 18 or older to use it. The feature supports eight languages including English, Spanish, French, German, Japanese, Korean, Hindi, and Portuguese. More languages will be added later.
All Gemini users get basic access. Google AI Plus, Pro, and Ultra subscribers get higher usage limits. The feature is also expanding to YouTube’s Dream Track for Shorts, allowing creators to generate custom background music for their videos.
Copyright and Safety
AI music undress ai tools has faced serious copyright challenges. In 2024, major record labels sued Suno and Udio for training their models on copyrighted songs without permission. Those lawsuits resulted in settlements and ongoing legal battles. Suno settled with Warner Music Group but remains in litigation with Sony and Universal. Udio settled with Universal and Warner but faced restrictions on downloads.
Google is taking a different ai porn video generator approach. The company emphasizes that Lyria 3 is trained on licensed data and original content. Google has also implemented strong safety measures.
Every track generated by Gemini carries a SynthID watermark. This is an invisible marker that survives compression, remixing, and format changes. If you upload an audio file to Gemini and ask if it was created by Google AI, the system can detect the watermark and tell you. This helps platforms like YouTube identify AI-generated content and avoid copyright disputes.
Google also states that Lyria 3 will not mimic specific artists. If you ask it to copy a famous singer’s voice, it will refuse. The model is designed to create original music rather than replicas.
Impact on the AI Music Market
Google’s entry into AI music generation changes the competitive landscape. Suno and Udio have been the leading players in this space. Suno raised 250 million dollars and has over 2 million paying subscribers. Udio has also grown quickly with strong backing.
However, Google has advantages that startups cannot match. Gemini has 750 million monthly active users. The music feature is free to use with basic limits. It is integrated into an app that people already use every day. For casual users who want to create a quick song for a birthday or social media post, Gemini is now the easiest option.
Industry observers note that while Gemini’s 30-second limit is shorter than Suno’s full-length songs, Google’s distribution power is unmatched. As one reviewer from Tom’s Guide wrote, Google is “shaking up the AI music space” by making music creation as easy as sending a text message.
The timing is also significant. Google is launching this feature while Suno and Udio are still dealing with legal uncertainty. By offering a tool with clear licensing, built-in watermarking, and direct YouTube integration, Google may attract users who are worried about copyright risks.
Limitations
Despite the excitement, there are some limitations to keep in mind.
The current beta version only generates 30-second clips. This is fine for social media and short videos but not enough for full songs. The quality, while impressive for AI, still shows some rough edges. Lyrics can sometimes be odd or repetitive. The feature is positioned as a creative tool rather than a professional production platform.
There is no API access yet. Developers cannot integrate Lyria 3 into their own apps programmatically. Google has hinted that API access may come later, but for now the feature is limited to the Gemini app.
What This Means for Creators
For content creators, musicians, and casual users, Gemini’s music feature opens new possibilities. You can now create original background music for videos, personalized songs for friends, or quick demos of musical ideas without any technical knowledge.
The integration with YouTube Dream Track is particularly useful for video creators. They can generate custom music for Shorts directly within the Google ecosystem, avoiding copyright claims and platform takedowns.
For professional musicians, the tool is more of a brainstorming aid than a replacement for human creativity. The 30-second limit and lack of professional mixing controls mean it cannot yet compete with full production software. However, as the technology improves, it could become a serious tool for rapid prototyping and creative exploration.
Final Thoughts
Google Gemini’s music generation feature is a significant step forward for AI creativity. By combining text, image, and video input with automatic lyrics and vocals, Google has created one of the most accessible music tools available.
The 30-second clips and beta status mean this is just the beginning. As Google expands the feature, adds API access, and increases track length, it could reshape how music is created and consumed.
For now, the biggest impact may be on the competitive landscape. With 750 million users, clear licensing, and built-in safety features, Google has set a new standard for AI music generation. Suno, Udio, and other competitors will need to adapt quickly to stay relevant.
When music creation becomes as easy as typing a sentence, everyone becomes a potential artist. Google’s bet is that the future of music is not just about listening. It is about creating, sharing, and expressing yourself in ways that were never possible before.
