Our Blog
01
Can Data Predict a Hit Song? Exploring Patterns with Machine Learning
Can data help predict a hit song? This project explores the intersection of music and machine learning by creating a model that predicts whether a song will become a hit based on its audio features (e.g., danceability, energy, loudness). The goal is to provide actionable insights for the music industry by leveraging data-driven methods.
02
Forecasting Spotify Listening Trends: A Data Science Project
Music shapes our everyday lives, and Spotify is a major part of that experience. In my latest project, I analyzed a Spotify user’s listening history to uncover behavioural patterns and predict future trends using time-series forecasting.
03
StemAI: AI-Powered Audio Stem Separation Using Deep Learning
StemAI is a user-friendly tool designed to simplify music stem separation. While several AI-powered stem separation tools exist, StemAI takes it a step further by analyzing the energy distribution of each extracted stem. This provides a unique visualization of how different musical elements interact within a song.
04
What Are the World’s Most Streamed Songs Really Saying? I Used GPT-4 to Find Out.
Today’s global music trends are more than just sound — they’re emotion, energy, and stories. I wanted to know: What are the dominant themes and moods driving the songs we stream most? So I built a Python tool to analyze the lyrics of the Spotify Top 50 Daily Global Chart — powered by OpenAI’s GPT-4 and Genius API.
