As AI slowly creeps into every facet of our digital lives — from writing essays to talking with therapists to generating original art — it’s surprising, to say the least, that artificial intelligence also take a step into the music. It’s also no surprise that tech giant Google is the first major player on the scene.
The company is reportedly built an AI bot that can create “original” music(Opens in a new window) from text and sound prompts – users could type increasingly specific prompts noting genres and styles or even create songs based on a hummed or whistled melody, according to reports. The future application is known internally as MusicLM.
The information was revealed in a research paper(Opens in a new window) published on January 26, describing MusicLM as a “model generating high-fidelity music from text descriptions” that “generates music at 24 kHz that stays consistent over several minutes”. The document explains that songs can be created from richly written captions, such as:
The soundtrack of an arcade game. It’s fast and upbeat, with a catchy electric guitar riff. The music is repetitive and easy to remember, but with unexpected sounds, like crashing cymbals or drum rolls.
Additional sequences of timed text prompts help build song structure(Opens in a new window)as well as a library of sounds and other AI prompts generated from sources such as art archives.
Examples of AI-generated songs have already been posted on Google’s Github account,(Opens in a new window) part of a preliminary version of a dataset of 5,500 music-text pairs known as MusicCaps.
The unveiling of such a platform will inevitably spark new conversations about the role of artificial intelligence in intellectual property theft and copyright infringement, generated by a plethora of artists and art deposits that have not consented(Opens in a new window) to the public use of their art in creating AI robots like these – others, meanwhile, are capitalizing on the rise of AI-based technology. Additional AI developments also present unique risks to the humans behind the technology, as the exploited workforce faces the brunt of data mining and moderation.
For now, the Google AI music maker won’t be released anytime soon, with the company explaining lingering concerns about cultural programming bias, glitches, and plagiarism concerns that need to be addressed before it launches.