Predicting Music Track Popularity by Convolutional Neural Networks on Spotify Features and Spectrogram of Audio Waveform

TL;DR

This work tackles the challenge of predicting music track popularity in the digital streaming era by combining audio-derived Mel spectrogram features with Spotify metadata in a convolutional neural network framework. The multimodal CNN is trained on a diverse US dataset (~2000 tracks across 100 artists) and evaluated with an 80/20 train/test split, employing early stopping to enhance generalization. The model achieves strong performance (approximately 95.68% accuracy and MAE around 9.50) and outperforms several baselines, with artist-related features showing especially strong predictive power. The study provides actionable insights for artists and industry stakeholders and sets a foundation for predictive analytics in music marketing and release planning.

Abstract

In the digital streaming landscape, it's becoming increasingly challenging for artists and industry experts to predict the success of music tracks. This study introduces a pioneering methodology that uses Convolutional Neural Networks (CNNs) and Spotify data analysis to forecast the popularity of music tracks. Our approach takes advantage of Spotify's wide range of features, including acoustic attributes based on the spectrogram of audio waveform, metadata, and user engagement metrics, to capture the complex patterns and relationships that influence a track's popularity. Using a large dataset covering various genres and demographics, our CNN-based model shows impressive effectiveness in predicting the popularity of music tracks. Additionally, we've conducted extensive experiments to assess the strength and adaptability of our model across different musical styles and time periods, with promising results yielding a 97\% F1 score. Our study not only offers valuable insights into the dynamic landscape of digital music consumption but also provides the music industry with advanced predictive tools for assessing and predicting the success of music tracks.

Figures (6)

• Figure 1: Flow of Song Popularity Prediction Using Spotify Data and CNN.
• Figure 2: The validation error plotted across training epochs, illustrating the model's improvement and convergence over time.
• Figure 3: Evolution of Mean Absolute Error (MAE) throughout the training epochs.
• Figure 4: A bar chart comparison of predicted versus actual popularity scores, highlighting the model's performance and accuracy in predictions.
• Figure 5: Histogram of Prediction Errors, indicating the frequency distribution of errors in a prediction model. The vertical dashed line represents the mean error.