From 78848a0984a9fbd23f9e03a566e04a39451ef581 Mon Sep 17 00:00:00 2001 From: Sevag Hanssian Date: Sun, 12 Mar 2023 15:45:00 -0400 Subject: [PATCH] Add arXiv badge --- README.md | 2 ++ 1 file changed, 2 insertions(+) diff --git a/README.md b/README.md index d4f1008..950da25 100644 --- a/README.md +++ b/README.md @@ -1,5 +1,7 @@ # xumx-sliCQ +[![arXiv](https://img.shields.io/badge/arXiv-2112.05509-b31b1b.svg)](https://arxiv.org/abs/2112.05509) + A variant of [Open-Unmix](https://github.com/sigsep/open-unmix-pytorch) that replaces the STFT with the [sliCQ transform](https://github.com/sevagh/nsgt) ([Balazs et al. 2011](http://ltfat.org/notes/ltfatnote018.pdf) and [Holighaus et al. 2012](https://arxiv.org/abs/1210.0084)) and uses convolutional layers (based loosely on [Grais, Zhao, and Plumbley 2019](https://arxiv.org/abs/1910.09266)). Transforms with nonuniform frequency spacing can better represent the tonal and transient characteristics of musical signals, such as the CQT. The sliCQ transform is one such transform: