Can an artificial intelligence learn to compose music by listening to 20,000 hours?

New AI Listened To 20,000 Hours Of Music. What Did It Learn? · 7:33 · Video: Two Minute Papers · Watch at the source

In short: This video examines MusicGen, a system trained on a massive dataset of audio to generate and control musical compositions. It highlights the technical progress in AI-driven audio synthesis and the potential for creating complex, controllable soundscapes.

The research, detailed in the paper 'MusicGen: Simple and Controllable Music Generation,' explores how large-scale training data can be leveraged to produce high-quality musical output. By processing 20,000 hours of music, the model learns to understand and replicate intricate patterns, allowing users to guide the generation process through specific controls.

This development represents a significant step in generative AI, moving beyond simple mimicry toward a more nuanced, controllable form of creative synthesis. The ability to steer the output suggests a future where AI acts as a sophisticated tool for musicians and composers, rather than just a passive generator.

Opens the Intellect+ feed with this card first. Swipe up for the next one.

Sources

To cite this card, link to https://intellectplus.co.uk/app/technology-new-ai-listened-to-20-000-hours-of-music-what-did-it-learn.

How we source, check and correct cards: Editorial standards.

General information, not professional advice. Spotted an error? Flag the card or see corrections and complaints.

Intellect Plus is a visual educational discovery platform for exploring fascinating facts and ideas. Brain Food, Not Brain Rot. What is Intellect Plus?

Related

Finding something worth knowing…