Views
No views yet
| model | si-sdr | pesq-wb | stoi-ext | lsd | lps | OVRL | SIG | BAK |
|---|---|---|---|---|---|---|---|---|
| UNIVERSE++ | 18.624 | 3.017 | 0.864 | 4.867 | 0.937 | 3.200 | 3.489 | 4.040 |
| UNIVERSE | 17.600 | 2.830 | 0.844 | 6.318 | 0.920 | 3.157 | 3.457 | 4.013 |
open-universe.
We use conda to simplify the installation.1git clone https://github.com/line/open-universe.git
2cd open-universe
3conda env create -f environment.yaml
4conda activate open-universe
5python -m pip install .1# UNIVERSE++ (default model)
2python -m open_universe.bin.enhance <input/folder> <output/folder> \
3 --model line-corporation/open-universe:plusplus
4
5# UNIVERSE
6python -m open_universe.bin.enhance <input/folder> <output/folder> \
7 --model line-corporation/open-universe:original1@inproceedings{universepp,
2 authors={Scheibler, Robin and Fujita, Yusuke and Shirahata, Yuma and Komatsu, Tatsuya},
3 title={Universal Score-based Speech Enhancement with High Content Preservation},
4 booktitle={Proc. Interspeech 2024},
5 month=sep,
6 year=2024
7}1@misc{universe,
2 authors={Serr\'a, Joan and Santiago, Pascual and Pons, Jordi and Araz, Oguz R. and Scaini, David},
3 title={Universal Speech Enhancement with Score-based Diffusion},
4 howpublished={arXiv:2206.03065},
5 month=sep,
6 year=2022
7}