Resources for the paper: Self-Supervised VQ-VAE for One-Shot Music Style Transfer
By O. Cífka

In our paper (Cífka et al., 2021), we present a novel method for one-shot timbre style transfer, based on the vector-quantized variational autoencoder (VQ-VAE) and self-supervised learning.

Links

• Paper preprint
• Companion website with audio examples
• Source code on GitHub
• Interactive Colab notebook
• Presentation

The paper

Cífka, O., Ozerov, A., Şimşekli, U., & Richard, G. (2021, June). Self-Supervised VQ-VAE for One-Shot Music Style Transfer. ICASSP 2021 - IEEE International Conference on Acoustics, Speech and Signal Processing. https://doi.org/10.1109/ICASSP39728.2021.9414235