Improving singing voice separation with the Wave-U-Net using Minimum Hyperspherical Energy

Re-assign community