Examples for ICASSP2024 paper “StemGen: A music generation model that listens”.
Lyria by Google DeepMind
Google DeepMind describes a music generation model and experiments intended to explore new ways to create music.
About Lyria by Google DeepMind
Lyria is a music generation model developed by Google DeepMind in partnership with YouTube. It generates music with instrumentals and vocals, performs transformation and continuation tasks, and allows control over the style and performance of the output.
It is being trialed through Dream Track in YouTube Shorts and through music AI tools developed with participants in YouTube's Music AI Incubator. Dream Track is limited to a limited set of creators and generates 30-second soundtracks, and output published by the Lyria model is watermarked with SynthID.
Key features
- Generate music with instrumentals and vocals
- Perform music transformation and continuation tasks
- Adjust output style and performance
- Generate 30-second Shorts soundtrack from topic and artist style
- Simultaneously generate lyrics, backing track, and AI-generated voice
- Create instrumental and vocal accompaniments
Best for: Artists, songwriters, producers, and Shorts creators involved in YouTube's music experiments.
Good for
- Allows control over output style and performance
- Generates lyrics, backing track, and voice together in Dream Track
- Marks published output with inaudible SynthID watermark
Watch out for
- Dream Track is available only to a limited set of creators
- Dream Track soundtracks are limited to 30 seconds
- Music AI tools are still being tested with incubator participants
Lyria by Google DeepMind at a glance
- Category
- AI & Machine Learning › AI Music Generation
- Type
- Website
- Pricing
- Not confirmed. Check the site
- Platforms
- Web
- Open source
- Not stated
- Language
- English
- Last checked
- Oct 5, 2026 · Online when checked
- Listed
- Oct 5, 2026
Lyria by Google DeepMind alternatives
Tools and sites you could use instead, ranked by how closely they match.
MusicLM explores ways to generate music based on written descriptions.
Google Flow Music supports creating, remixing, and sharing songs with generative AI.
ACE-Step is a local music generation model supporting Mac, AMD, Intel, and CUDA devices.
Open source
24 alternatives
Stable Audio provides customizable models and tools for iterative music and sound creation.
This AI song maker turns text and lyrics into generated music tracks for users to explore.
Popular in AI Music Generation
Well-known listings from the same topic.
Suno AI Musicsunoai-music.comEcrett Musicecrettmusic.comMusicCreator AImusiccreator.aiMubertmubert.comAudioXaudiox.appAIVAaiva.ai
Lyria by Google DeepMind FAQ
What does Dream Track do?
Users enter a topic and choose a participating artist to generate a 30-second soundtrack for a Short, including lyrics, backing track, and AI-generated voice.
Which artists are participating in Dream Track?
Alec Benjamin, Charlie Puth, Charli XCX, Demi Lovato, John Legend, Sia, T-Pain, Troye Sivan, and Papoose.
How is audio generated by Lyria identified?
Content published by the Lyria model is watermarked with SynthID, which embeds a watermark that is inaudible and does not compromise the listening experience.
Is something wrong or out of date?
Tell us and an editor will check it.
Own this website?
Claim this listing
Claim the listing and suggest updates yourself. An editor checks every claim first.