Meta* has introduced a new generative AI MusicGen, which can compose music from a textual description. about this corporation informed On twitter.
The artificial intelligence was developed by Audiocraft’s in-house team, at this stage MusicGen creates compositions up to 12 seconds long.
According to the developers, the AI was trained for 20 thousand hours of music. 10,000 licensed songs and 390,000 free tracks from the Shutterstock and Pond5 libraries were used for this.
To get a melody, you need to enter a short text request with a description of the composition. You can also tell the AI what kind of melody to compose by uploading your own sample song.
Unlike Google’s Music AI (MusicLM), the Meta* model is not capable of producing vocals.
Now access to MusicGen is open to everyone. You can test the model on the Hugging Face website. Some experts note that the model can attract attention in the context of copyright protection, since it was trained on licensed tracks.
Previously, Meta* introduced a new and more advanced Megabyte generative modeling framework.
*The activity of Meta Platforms is recognized as extremist on the territory of the Russian Federation and is prohibited.
Author:
Natalia Gormaleva
Source: RB

I am a professional journalist and content creator with extensive experience writing for news websites. I currently work as an author at Gadget Onus, where I specialize in covering hot news topics. My written pieces have been published on some of the biggest media outlets around the world, including The Guardian and BBC News.