DeepFilterNet: Perceptually Motivated Real-Time Speech Enhancement

Hendrik Schröter, Tobias Rosenkranz, Alberto N. Escalante-B., Andreas Maier
5/14/2023
eess.AScs.CLcs.SD

No official code found. Best implementation: Rikorose/DeepFilterNet (4,622 stars, PyTorch, last commit 2024-10-17).

Abstract

Multi-frame algorithms for single-channel speech enhancement are able to take advantage from short-time correlations within the speech signal. Deep Filtering (DF) was proposed to directly estimate a complex filter in frequency domain to take advantage of these correlations. In this work, we present a real-time speech enhancement demo using DeepFilterNet. DeepFilterNet's efficiency is enabled by exploiting domain knowledge of speech production and psychoacoustic perception. Our model is able to match state-of-the-art speech enhancement benchmarks while achieving a real-time-factor of 0.19 on a single threaded notebook CPU. The framework as well as pretrained weights have been published under an open source license.

View on arXivView PDF

Code Implementations(1)

Rikorose/DeepFilterNetCommunity100%
PyTorch

Noise supression using deep filtering

4,622504Rust, ShellOct 10, 20212 years agoNOASSERTION
audiodeep-learningnoise-suppressionpytorchrust+2 more

Similar Papers

RT-LA-VocE: Real-Time Low-SNR Audio-Visual Speech Enhancement

83% match2024

FINALLY: fast and universal speech enhancement with studio-like quality

81% match2026

xLSTM-SENet: xLSTM for Single-Channel Speech Enhancement

81% match1 repo602025

PAAPLoss: A Phonetic-Aligned Acoustic Parameter Loss for Speech Enhancement

80% match1 repo312023

EffiFusion-GAN: Efficient Fusion Generative Adversarial Network for Speech Enhancement

80% match2025

When AVSR Meets Video Conferencing: Dataset, Degradation, and the Hidden Mechanism Behind Performance Collapse

80% match2026

Cite this paper

@article{schrter2023deepfilternet,
  title  = {DeepFilterNet: Perceptually Motivated Real-Time Speech Enhancement},
  author = {Hendrik Schröter and Tobias Rosenkranz and Alberto N. Escalante-B. and Andreas Maier},
  year   = {2023},
  eprint = {2305.08227},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2305.08227v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2305.08227.svg)](https://codeofpaper.com/paper/2305.08227)

Discussion