A study detailed in the Frontiers in Artificial Intelligence journal claims to have accurately predicted hit songs with 97% accuracy. Researchers took a new approach to predicting hit songs by directly measuring neurophysiologic responses to songs rather than relying on subjective like and dislike models. They used a machine learning model trained on that data that can predict hit songs with high accuracy.
Implications for the Music Industry
The study raises questions about how this technology will impact the music industry. How will labels and distributors use a model like this to find talent and manage their investments? Will this narrow the music industry or widen it to new possibilities? The study says their model “provides a new way for artists, record producers, and streaming services to delight listeners with new music.”
A possible implementation method they describe is “wearable neural [technology]” that could assess the listener’s reaction to songs automatically and help streaming services curate new music specific to the listener. As biometric sensors become more readily available on personal wearables like smartwatches, could we see streaming services utilize these sensors to curate music?
It remains to be seen how this technology will be used in practice, but it is clear that it has the potential to revolutionize the way the music industry operates.
To learn more, read the study here.