---
license: mit
---

This is a pre-trained version of Fast FullSubNet, trained on the Deep Noise Suppression Challenge dataset of 2020 (DNS-INTERSPEECH-2020).

# Instructions

https://fullsubnet.readthedocs.io/en/latest/usage/getting_started.html

# Code

https://github.com/Audio-WestlakeU/FullSubNet

# Paper

[Fast FullSubNet: Accelerate Full-band and Sub-band Fusion Model for Single-channel Speech Enhancement
Xiang Hao, Xiaofei Li](https://arxiv.org/abs/2212.09019)

# Performance

|   | With Reverb |   |   |   | No Reverb |   |   |  
-- | -- | -- | -- | -- | -- | -- | --
Method                 | WB-PESQ | NB-PESQ | SI-SDR | STOI | WB-PESQ | NB-PESQ | SI-SDR | STOI
Fast FullSubNet (118 Epochs) | 2.882 | 3.42 | 15.33 | 0.9233 | 2.694 | 3.222 | 16.34 | 0.9571
[FullSubNet (58 Epochs)](https://github.com/Audio-WestlakeU/FullSubNet/releases/tag/v0.2) (just for comparison) | 2.987 | 3.496 | 15.756 | 0.926 | 2.889 | 3.385 | 17.635 | 0.964