Juke box jury computer program 'predicts hit songs'
Record companies are to test a "juke box jury" computer program that can tell if a song is likely to top the charts.
Like the guests on the TV pop show, launched in the 1960s and revived in the 1980s, the software predicts whether a tune is going to be a "hit" or a "miss".
The program, called Hit Song Science, correctly forecast the success of jazz songstress Norah Jones months before she topped the US charts and won eight Grammy awards.
Record company bosses were so impressed that five major labels have decided to try out HSS for themselves, New Scientist magazine reports.
HSS, produced by the Spanish company Polyphonic HMI, of Barcelona, looks for songs that match the musical traits of known hits.
Each song is run through a set of signal filters that identify and measure more than a dozen musical patterns, including melody, harmonic variation, beat, tempo, rhythm, pitch, chord progression and fullness of sound.
The program's designers found that in the past five years of Billboard magazine's Top 30 chart listings, hits were concentrated into a number of small clusters sharing similar traits.
Polyphonic HMI's chief executive Mike McCready said: "There are a limited number of mathematical formulas for hit songs. We don't know why."
Peter Bentley, of University College London, who also designs musical software, said it may only reinforce tried and trusted formulas for success.
"The music industry is not exactly renowned for its daring exploits," he said. "If you rely on the computer too heavily, you will miss out on the new things."