Software developed by MIT teaches computers to predict the next music hits, the Guardian reports. The computers are trained to hear music much like people do. This means analysing the...

Software developed by MIT teaches computers to predict the next music hits, the Guardian reports. The computers are trained to hear music much like people do. This means analysing the context of the song – what people think of it – as well as how fast or slow it is, and whether it is loud or soft. The result is a sophisticated profile of a song that allows more accurate categorisation than simply labelling a song as rock, pop or classical.

"The computer can go out on the internet and read a bunch of weblogs and record reviews and get a grasp on trends and buzz and hype and that sort of thing," says Dr Whitman. "At the same time, it also can listen to the music – do some signal processing to figure out stuff like dominant key and pitch and rhythm and structure….. For record company executives, this raises the tantalising possibility of knowing in advance whether their latest pop act will hit the charts at a strong position."

READ THIS ARTICLE FOR $15
$15 provides access to this article and every case-study, interview, and analysis piece that we publish for the next 30 days. Our Premium Subscription also provides access to a database of over 100,000 articles on innovation in brand, customer, and retail experience.
Already a subscriber? Log in