
Share
Explore how Google's MusicLM transforms descriptions into unique soundscapes, blending over 100 instruments to create abstract clips that defy traditional music boundaries.
Google has launched a new generative AI experiment called Instrument Playground, which allows users to create abstract musical clips inspired by over 100 instruments from around the world. This tool, designed by Simon Doury, an Artist in Residence at Google Arts & Culture Lab, leverages Google’s MusicLM, a text-to-music AI model that became publicly available in May.
Instrument Playground starts by asking for a simple prompt containing the name of a musical instrument, optionally preceded by an adjective like "upbeat," "strange," or "gloomy." The system then generates a 20-second audio clip as a starting point for your composition. These clips are often abstract and may not include the specific sound of the instrument you entered but rather capture the essence or feeling of the prompt.
Once you have a clip you like, you can further refine it using the following options:

For more complex compositions, an advanced mode opens up a sequencer to layer and loop up to four tracks. This allows you to create a more unique and personalized musical piece. Once you’re satisfied, you can download your creation as a .wav file.
Google has provided a holiday-themed example to inspire users. If you’re ready to dive in, visit Instrument Playground and log in with your Google account to start composing.
Instrument Playground offers a unique and creative way to explore the intersection of AI and music. Whether you’re a seasoned musician or just curious about what an "angry tuba" might sound like, this tool provides a fun and accessible entry point into generative music.
Tags
Original Sources
About the author
Kai built ML infrastructure at a Bay Area startup before developing an obsession with transformer architectures and inference optimisation that eventually pulled him out of product work entirely. A stint at a compute research lab sharpened his instinct for what actually matters in a model release versus what is marketing. He writes from the inside — from the perspective of someone who has debugged the systems he is describing at three in the morning. He is allergic to hype and instinctively drawn to the unglamorous plumbing questions that everyone else skips over.
More from The Engineer →This Week's Edition
6 December 2023
88 articles
Related Articles

OpenEvidence Targets Hospitals to Expand Its AI Chatbot for Doctors
Products & Applications · 3 min

OpenEvidence Launches Voice AI to Enhance Physician Workflow
Products & Applications · 3 min

Doximity Accelerates AI Investment in 2026, Targeting Multibillion-Dollar Market
Products & Applications · 3 min
Related Articles

OpenEvidence Targets Hospitals to Expand Its AI Chatbot for Doctors
Products & Applications · 3 min

OpenEvidence Launches Voice AI to Enhance Physician Workflow
Products & Applications · 3 min

Doximity Accelerates AI Investment in 2026, Targeting Multibillion-Dollar Market
Products & Applications · 3 min
More Stories