factual

AI engine learns sweet spots to predict music preference

By extracting seven musical features (harmony, melody, rhythm, timbre, texture, dynamics) at scale via music information retrieval from thousands of songs' audio signal, an AI engine can learn the context-specific weighting and 'sweet spots' that accounted for variance in past preferences and apply them to predict reception of new songs.

factualpending

Speaker

Scott Miles

Evidence Quote

“these are the sweet spots that the algorithm and the artificial intelligence through the machine learning will find and apply to get the best predictions”

Source

104 | David Rosen and Scott Miles on the Neuroscience of Music and Creativity— Sean Carroll's Mindscape
Created: 6/13/2026, 12:22:23 AM

Your notes

Loading notes...