Google Unveils MusicLM: An AI-Powered Music Generation Software
Google has developed a new music generation software, MusicLM, which is capable of producing music in any genre using text descriptions. Despite the software not being released yet, the development has generated a lot of excitement, as well as concerns over the potential risks associated with AI and music creation.
MusicLM: Trained on a Dataset of 280,000 Hours of Music
MusicLM was trained on a dataset of 280,000 hours of music, which allowed it to learn and generate music from descriptions. The authors of the research paper on MusicLM acknowledged the risks of creative content misappropriation and emphasized the need for future work to address these risks.
Optimism in the Music and AI Industries
Experts in the music and AI industries are optimistic about the potential for MusicLM to revolutionize the song creation process. Martin Clancy, a musician and AI expert, sees the capability for users to select their preferred genre, adjust for their music skill level, and even tailor the music to specific physical settings. CEO of Beatoven.ai, Mansoor Rahimat Khan, highlights the potential for AI to aid in generating musical ideas and pushing beyond the limitations of an artist's own creativity.
Concerns Over Homogenization of Music
However, there are also concerns that AI could lead to the homogenization of music and the loss of value placed on the skills of music composition and production. DJ and AI enthusiast, SPLY85, comments that creating music with AI requires a deep understanding of both music theory and programming.
AI as a Tool for Enhancing Creativity
Music producer Emilio Guarino, who incorporates AI in his work, has experienced unexpected outcomes from the software and sees AI as a way to enhance creativity by providing novel stimulus. Despite this, Guarino and others do not predict AI completely replacing humans in the musical process, but rather being limited to "wallpaper music" for background use.