A new AI system from Google called “MusicLM” can produce high-fidelity music in any genre from a text description.
Per the TechCrunch, the company is afraid of the risks and does not have any imminent plans to release it.
Google isn’t the first to experiment with generative AI for songs; other efforts have included Dance Diffusion, OpenAI’s Jukebox, and Riffusion, an AI that creates music by visualizing it.
However, none of them has been able to create songs that are exceptionally sophisticated in composition or high quality due to technical limitations and a lack of training data.
Yesterday, Google published a paper on a new AI model called MusicLM.— Product Hunt 😸 (@ProductHunt) January 27, 2023
The model generates 24 kHz music from rich captions like "A fusion of reggaeton and electronic dance music, with a spacey, otherworldly sound. Induces the experience of being lost in space." pic.twitter.com/XPv0PEQbUh
The report said that MusicLM is maybe the first to be able to.
Using a dataset of 2,80,000 hours of music, MusicLM was trained to create songs that were cogent for descriptions of “significant complexity,” such as “enchanting jazz song with a memorable saxophone solo and a solo singer” or “Berlin ’90s techno with a low bass and strong kick,” as per a Cornell University academic paper.