A new AI-powered service helping music companies, music scientists and musicians to make better informed decisions whenever a binary or multi-class prediction or classification is required. For example:
1. Predict song popularity: Having extracted and recorded the data features of thousands of past songs including artist, song writer, year of release, month of release, catchy melody (yes or no), easy to remember lyrics (yes or no), repeated chorus (yes or no), predictability score within the first quarter (within song duration), predictability score within the second quarter, predictability score within the third quarter, predictability score within the last quarter, danceability, energy level within the first quarter of the of song, energy within the second quarter, energy within the third quarter, energy within the fourth quarter, key, loudness, mode, speech-ness, acoustic-ness, instrumental-ness, lively-ness, valence, tempo, duration, time signature, class of music (Jazz, country, hip hop, rock, folk music, etc.), and having verified and labelled each one of those past songs as (not popular: below 0.3333, popular: between 0.3333 and 0.6666, and a hit: above 0.6666, labels and target of prediction), predict the level of popularity of a new song provided that the data features, described above, are available for that song.