Lyria 2 delivers high-fidelity music and professional-grade audio, capturing subtle nuances across a range of genres and intricate compositions.

Model family

Lyria is a family of models for music generation providing capabilities for both offline and real-time music generation.

music_note

Lyria 2

Designed to produce high-quality music primarily through offline generation.

youtube_live

Magenta RealTime

Open-source Lyria model explores the role of machine learning as a tool in the creative process.


Capabilities

Powerful composition, endless exploration

Lyria 2 is an advanced AI model, developed with input and insights from musicians and producers. Artists can use text prompts to shape the music, while controlling the key, BPM, and other characteristics.

high_quality

High-fidelity music

Generates music with remarkable detail and richness, capturing nuance across different instruments and playing styles. Produces professional-grade 48kHz stereo audio, ready for seamless integration into any project.

instant_mix

Granular creative control

Allows musicians to control fine details within their compositions, so they can create pieces that match their intent and vision.

music_note

Diverse musical possibilities

Allows composition across a wide range of genres and styles. Classical, jazz, pop, electronic — musicians can steer compositions towards their tastes, and explore diverse musical ideas along the way.



Your creative partner

edit_audio

Keep the music flowing

Lyria 2 can create new starting points, suggest harmonies, or draft longer arrangements. It helps musicians focus on their broader visions — or jump past writer’s block.

slow_motion_video

Accelerate creativity

Composing complex passages and creating variations on themes can take hours, or even days. Lyria 2 accelerates the process, allowing musicians to bring their musical ideas to life in no time.

queue_music

Uncover new styles

Discover unfamiliar genres, techniques, and possibilities. Experiment with unexpected melodies and harmonies. Lyria 2 can broaden musical understanding and help people explore new creative avenues.


Music AI Sandbox

Powered by Lyria 2, Music AI Sandbox is a tool we’ve developed with insights from music industry professionals.

spatial_tracking

A musical playground

Create new instrumental or vocal parts for existing pieces, or explore new directions — amplifying creativity. Music AI Sandbox helps artists push into unexplored territory, from unique soundscapes to songs that take musicality to its limits.

groups

Built with, and for, musicians

We created Music AI Sandbox in collaboration with artists. Their input has guided our development, resulting in a set of responsibly-created tools that are practically useful for artists of all levels — opening doors to music creation.



Limitations

While Lyria 2 demonstrates the incredible potential of AI for music generation, we are still working to develop and refine its performance — and to expand its coverage of languages and genres.


Safety

From development to deployment

We believe AI should enhance human creativity, not replace it. We’re working directly with artists to gather feedback and to develop guardrails that address their concerns. By doing this, we ensure this tool empowers creators and helps them realize the possibilities of AI in their work.

We’re also deploying Lyria 2 with our latest privacy, safety and security technologies, including our innovative watermarking tool SynthID — which embeds an imperceptible digital watermark directly into the audio output, making it identifiable as AI-generated content but otherwise imperceptible.


Try Lyria


Acknowledgements

Lyria 2 was developed by Andrea Agostinelli, Asahi Ushio, Beat Gfeller, Brian McWilliams, Ethan Manilow, Kazuya Kawakami, Keyang Xu, Matej Kastelic, Mauricio Zuluaga, Mauro Verzetti, Myriam Hamed Torres, Ondrej Skopek, Pavel Khrushkov, Pen Li, Tobenna Peter Igwe, and Zalan Borsos. Additional contributions provided by Adam Roberts,, Benigno Uria, Carrie Zhang, Chris Deaner, Colin McArdell, DY Kim, Eleni Shaw, Hongliang Fei, Jason Baldridge, Jesse Engel, Li Li, Luyu Wang, Nemanja Spasojevic, Noah Constant, Ruba Haroun, Tayniat Khan, Volodymyr Mnih, Yan Wu and Zoe Ashwood.