Exploiting Synchronized Lyrics And Vocal Features For Music Emotion Detection

TL;DR

This paper introduces a novel dataset and method for music emotion detection by combining synchronized lyrics and vocal audio features, demonstrating that text-based models can outperform audio-based ones in classifying music emotions.

Contribution

The work presents the Synchronized Lyrics Emotion Dataset and compares text and audio models, showing that text-based deep learning models with attention outperform audio classifiers.

Findings

Vocal-only audio improves emotion classification performance.

Bilinear LSTM with attention and fastText embeddings outperforms CNN on audio.

Synchronized lyrics enhance emotion detection accuracy.

Abstract

One of the key points in music recommendation is authoring engaging playlists according to sentiment and emotions. While previous works were mostly based on audio for music discovery and playlists generation, we take advantage of our synchronized lyrics dataset to combine text representations and music features in a novel way; we therefore introduce the Synchronized Lyrics Emotion Dataset. Unlike other approaches that randomly exploited the audio samples and the whole text, our data is split according to the temporal information provided by the synchronization between lyrics and audio. This work shows a comparison between text-based and audio-based deep learning classification models using different techniques from Natural Language Processing and Music Information Retrieval domains. From the experiments on audio we conclude that using vocals only, instead of the whole audio data improves the overall performances of the audio classifier. In the lyrics experiments we exploit the state-of-the-art word representations applied to the main Deep Learning architectures available in literature. In our benchmarks the results show how the Bilinear LSTM classifier with Attention based on fastText word embedding performs better than the CNN applied on audio.