Paper ID: 2312.09207

WikiMuTe: A web-sourced dataset of semantic descriptions for music audio

Benno Weck, Holger Kirchhoff, Peter Grosche, Xavier Serra

Multi-modal deep learning techniques for matching free-form text with music have shown promising results in the field of Music Information Retrieval (MIR). Prior work is often based on large proprietary data while publicly available datasets are few and small in size. In this study, we present WikiMuTe, a new and open dataset containing rich semantic descriptions of music. The data is sourced from Wikipedia's rich catalogue of articles covering musical works. Using a dedicated text-mining pipeline, we extract both long and short-form descriptions covering a wide range of topics related to music content such as genre, style, mood, instrumentation, and tempo. To show the use of this data, we train a model that jointly learns text and audio representations and performs cross-modal retrieval. The model is evaluated on two tasks: tag-based music retrieval and music auto-tagging. The results show that while our approach has state-of-the-art performance on multiple tasks, but still observe a difference in performance depending on the data used for training.

Submitted: Dec 14, 2023