https://arstechnica.com/information-technology/2023/01/googles-new-ai-model-creates-songs-from-text-descriptions-of-moods-sounds/ > On Thursday, researchers from Google announced a new generative AI model called MusicLM that can create 24 KHz musical audio from text descriptions, such as "a calming violin melody backed by a distorted guitar riff." It can also transform a hummed melody into a different musical style and output high-fidelity, sustained music for several minutes.