Google has now integrated Lyria 3, its latest music generation model, directly into the Gemini app, and you can start using it right away. Users simply tell it what they want or upload a photo, and within a few minutes, a 30-second single appears, complete with appropriate instruments, singers, and lyrics.
The model can handle a wide range of styles, such as when someone requests a high-energy Afrobeat piece inspired by family memories, and Lyria 3 creates something both personal and catchy. Another user uploads a photo from a vacation, and the system generates audio that matches the tone of the setting, possibly including lyrics that pick up on the feeling conveyed in the photo. Even instrumental works sound excellent when there is no singing involved. If the prompt does not have lyrics, it will just write them based on the mood, tempo, or data you provide.
Sale
Google Pixel 10a – Fog – 128 GB with $100 Amazon Gift Card
- Order the new Google Pixel 10a today and get an Amazon Gift Card; valid 2/18/2026 until 3/11/2026 at 11:59pm PT, while supplies last and subject to…
- Return of Pixel 10a without gift card results in charge; offer cannot be combined, is non-transferable and not valid for cash or cash equivalent
- If a qualifying item in your order is returned, you’ll be reimbursed for what you return, minus the value of the gift card
Control extends far beyond the initial prompt, as users can adjust things like drumming patterns or general tempo once the first version appears. You can also remix existing tracks by starting with something you already have and then experimenting with the pieces until they evolve into something new. Google couples each product with some custom cover art generated by its Nano Banana model, so the finished item is available to share on social media or download for personal use immediately.
The quality is undoubtedly higher than in previous attempts by the same team, as the tracks sound more natural, with smoother transitions between notes and richer layering that just avoids the robotic vibe from before. It can handle more complex tasks without losing its hold, whether it’s aiming for something simple and catchy or something more deep and nuanced. Every clip bears a Synth ID stamp as a clear indication that it is AI-generated, allowing users to determine where it came from if necessary.
For the time being, access remains very straightforward because the feature is currently in beta within the Gemini app and web version. Anyone 18 or older and fluent in a few languages (English, Spanish, German, French, Hindi, Japanese, Korean, and Portuguese) can start using it right immediately, but Google is limiting the outputs to 30-second portions to avoid having a half-finished work. They also have certain controls in place; if you try to mention a real artist in the prompt, it will avoid direct copying and instead suggest something with a similar vibe or mood.
It’s more than just a chatbot feature; YouTube producers in select locations can use a variation of the technology via the Dream Track tool to generate supporting audio for Shorts. That simply broadens the pool of people who can try their hand at music and provides some really powerful tools to those who have never picked up an instrument before.
