Spotify has filed a patent for a new plagiarism detection tool that has been designed to expose songs that copy aspects of other compositions, helping songwriters avoid lawsuits.
Music Business Worldwide reports that with the Plagiarism Risk Detector and Interface technology, a song’s lead sheet—a type of musical notation that indicates melody, chords, and lyrics—would be fed into the AI software that is “trained on a plurality of preexisting encoded lead sheets” to determine if a “similarity value” appears in other compositions in the database.
The software’s graphic user interface “provides dynamic visual feedback in substantially real-time,” permitting musicians and composers to discover any existing plagiarism and adjust elements of their work as they’re arranging it. “Such a tool,” the patent filing says, “would allow artists to generate lead sheets more quickly and confidently by detecting and providing visual feedback as to whether any aspect of the work has a probability of being deemed plagiaristic.”
Spotify alleges that the technology—which was created with the help of AI expert François Pachet—is much more intuitive and precise than extant methods like the Music Plagiarism Detection System, which is built on melodic similarity, and Audio Forensics Meets Music Information Retrieval – A Toolbox for Inspection Of Music Plagiarism, which is based on sampling, melody, and rhythm.
