Lyria 3.5
Create custom tracks for any moment
The new way to hit play. Experience Lyria 3.5, Gemini’s best-sounding AI music model. Transform text or photos into complete tracks with rich vocals and instrumentals. It’s never been easier to turn a specific feeling, photo, or mood into your perfect soundtrack.
Make it, hear it, share it.
Ask Gemini to turn a simple prompt into a brand-new sound that fits your style. Mix genres, moods and eras to make tracks to set the energy for a bit of focus, special occasions, a share-worthy anthem in the group chat and more.
New Sound. New controls.
Choose your sound, start with a template, and create longer, richer tracks with the latest from Lyria.
Vocal vs instrumental
Choose vocals or instrumentals to give your track its own groove.
Genre selection
Explore a wide range of genres to key in the exact sound and style you're looking for.
Track length up to 3 min
Give your ideas room to boogie with tracks up to three minutes long.
Start with a template
Dive into Lyria’s template gallery to find your perfect starting point. Select a track, tweak the details, and watch Gemini turn it into a high-fidelity soundtrack.
Express it with a track
Type a vibe. Get a beat. From inside jokes to your new personal anthem, describe what you want to hear and Gemini handles the rest to turn your prompt into a personalized, shareable track.
Soundtrack your camera roll
Turn your favorite meme or memory into shareable entertainment. Just upload a photo and Gemini will analyze the scene to compose a custom track that fits the feeling perfectly.
Frequently asked questions
Create high-quality tracks up to 3 minutes long with Lyria 3.5, our best sounding AI music model yet. Experience improved lyric quality and richer vocals, alongside enhanced creative controls that let you customize every track to reflect your personal style.
With the Gemini app, each track comes with custom cover art created by Nano Banana. Once your track is complete, you can easily download it or share it directly with friends across your favorite platforms.
AI music generation is available in all countries where the Gemini app is available. You must be 18+ to enjoy the experience.
Consistent with our AI principles, all tracks generated in the Gemini app are embedded with SynthID, our imperceptible watermark for identifying Google AI-generated content.
We are also giving you more tools to help identify AI content, broadening our verification capabilities to include audio. Simply upload a file and ask if it was generated using Google AI, and Gemini will check for SynthID and use its own reasoning to return a response.
To get the most out of Gemini's AI music generator, include these ingredients to refine your prompts and elevate your sound:
-
Genre and era: Lead with a specific genre, a unique mix or a musical era. (e.g. 80s synth-pop, metal and rap fusion, indie folk, old country)
-
Tempo and rhythm: Set the energy and describe how the beat feels. (e.g. Upbeat and danceable, slow ballad, driving beat)
-
Instruments: Ask for specific sounds or solos to add texture to your track. (e.g. Saxophone solo, distorted bassline, fuzzy guitars)
-
Vocals: Specify gender, voice texture (timbre) and range for the best delivery. (e.g. Airy female soprano, deep male baritone, raspy rocker)
-
Lyrics: Describe the topic, include personalised details or provide your own text with structure tags. (e.g. 'About an epic weekend' Custom: [Verse 1], A mantra-like repetition of a single word)
When you put it together, you can make something like this:
An indie folk track with a relaxed, swaying beat. The track features dry, intimate acoustic guitar, soft piano and light percussion. Soft, breathy female vocals sing lyrics about walking my dog on a cloudy day.
Limits apply. Compatibility varies. 18+. Create responsibly.