This is the repository for the code for the paper "Multimodal Raga Classification from Vocal Performances with Disentanglement and Contrastive Loss" by Sujoy Roychowdhury and Preeti Rao (Indian Institute of Technology Bombay, India) accepted for publication at Transactions of the International Society for Music Information Retrieval.
Update as of June 22, 2025: This repository is being updated with the latest codes for the paper.
The dataset used in this work is sourced from the hindustani_raga_dataset_processing GitHub repository.
The processed output data is available on the OSF project page. Specifically, we use the files located in the 04_masterfiles_3d directory under Files → OSF Storage.
Please refer to the GitHub repository for details on how the master files are generated from the raw videos.