Social media giant Meta has introduced MusicGen, its newest AI-powered music generator that utilizes text prompts to create music tracks that last up to 12 seconds.

Meta trained MusicGen on 20,000 hours of music, including 10,000 high-quality licensed tracks and 390,000 instrumental-only tracks from media libraries such as ShutterStock and Pond5. As a result, this powerful tool can generate music snippets based on a description or existing song references.

Users can experience MusicGen for free on the Hugging Face platform, while anyone with suitable hardware (primarily a GPU with about 16GB memory) can run the pre-trained models, which Meta has made available on GitHub.

Meta’s New AI Can Generate Short Videos From Text Prompts
The model analyses a text query and outputs 16 frames of video at a resolution of 64 by 64 pixels which are then boosted using a separate AI model to 768 by 768 pixels. Despite these limitations, Meta says that the resulting videos will become much better in the future.

Given its ability to generate tracks from descriptions like "An '80s driving pop song with heavy drums and synth pads in the background" or "acoustic folk song to play during road trips", MusicGen demonstrates promising potential for both music enthusiasts and professionals.