Paper Title
Enhanced Adaptive Equalization based on User Behavior using Clustering

Abstract
Enhancing the audio experience of the user has always been very imperative. Equalizers are used for adjustments of different frequencies of the audio components for the best and most personalized music experience for the user. But they are completely user-driven processes. There is no existing mechanism to suggest the best equalizer setting for a particular song or one that includes the taste of the user for a personalized experience. We propose a method that can customize equalizer settings based on learning user behavior and can further customize user playlists and music experiences. The proposed method first decodes themp3 file and further applies frequency analysis followed by the clustering of the songs, making similar songs lie in a single cluster, accounting for the taste of the user for best equalization. The model learns the user’s pattern of settings and does pattern recognition to learn the user’s taste. Further, we apply dynamic clustering according to the response of the user stored. We achieve the state of the result where the model can play the best songs on the best equalization settings and the model can learn the usage pattern based on the change of the equalizer settings in one of the songs in the cluster, hence making it extraconvenient to provide the best music experience