The metadata and pre-computed audio representations for the Fabric Dataset can be downloaded here: [https://drive.google.com/drive/folders/1dwUKaLRqAigRm2HwRWrAaSg4zKFdPnS6]
If you find this repository or dataset helpful, please cite the following paper:
Alexander Williams, Haokun Tian, Stefan Lattner, Mathieu Barthet, Charalampos Saitis; Deep Learning-based Audio Representations for the Analysis and Visualisation of Electronic Dance Music DJ Mixes. In AES International Symposium on AI and the Musician; 2024; Boston, MA, USA.
Alexander Williams and Haokun Tian are research students at the UKRI Centre for Doctoral Training in Artificial Intelligence and Music, supported jointly by UK Research and Innovation [grant number EP/S022694/1], Queen Mary University of London, and Sony CSL.
We wish to thank Fabric and the artists involved in the fabric and FABRICLIVE DJ mix series for their contribution to our dataset and Christopher Mitcheltree for helpful discussions on the JTFST.