It’s nice to see some good music-related submissions in ICLR 2019. Many are about timbre transfer as a part of music style transfer, which is great as now many are clarifying the scope of the problem rather than “style transfer” where the “style” is retrospectively defined by whatever happened in their neural networks.
Some of my comments are as below:
- “Autoencoder-based music translation”
- “TimbreTron: A WaveNet(CycleGAN(CQT(Audio))) Pipeline for musical timbre transfer”
Please go check out the submissions! → 11 results found for “music”