Speech Denoising
27 papers with code • 2 benchmarks • 3 datasets
Obtain the clean speech of the target speaker by suppressing the background noise.
Benchmarks
These leaderboards are used to track progress in Speech Denoising
| Trend | Dataset | Best Model | Paper | Code | Compare |
|---|---|---|---|---|---|
|
|||||
|
EmoSpeech
WHAMR_ext
Most implemented papers
Speech Denoising with Deep Feature Losses
anicolson/DeepXi
•
•
We present an end-to-end deep learning approach to denoising speech signals by processing the raw waveform directly.
WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing
microsoft/unilm
•
•
Self-supervised learning (SSL) achieves great success in speech recognition, while limited exploration has been attempted for other speech processing tasks.
Joint Optimization of Masks and Deep Recurrent Neural Networks for Monaural Source Separation
bill9800/speech_separation
•
•
In this paper, we explore joint optimization of masking functions and deep recurrent neural networks for monaural source separation tasks, including monaural speech separation, monaural singing voice separation, and speech denoising.
Speech Denoising Convolutional Neural Network trained with Deep Feature Losses.
francoisgermain/SpeechDenoisingWithDeepFeatureLosses
•
•
Interspeech 2018
We present an end-to-end deep learning approach to denoising speech signals by processing the raw waveform directly.
Speech Denoising Without Clean Training Data: A Noise2Noise Approach
madhavmk/Noise2Noise-audio_denoising_without_clean_training_data
•
•
This paper tackles the problem of the heavy dependence of clean speech data required by deep learning based audio-denoising methods by showing that it is possible to train deep speech denoising networks using only noisy speech samples.
FRA-RIR: Fast Random Approximation of the Image-source Method
yluo42/fra-rir
•
•
The training of modern speech processing systems often requires a large amount of simulated room impulse response (RIR) data in order to allow the systems to generalize well in real-world, reverberant environments.
Supervised and Unsupervised Speech Enhancement Using Nonnegative Matrix Factorization
We propose a novel speech enhancement method that is based on a Bayesian formulation of NMF (BNMF).
Investigating the effect of residual and highway connections in speech enhancement models
jfsantos/irasl2018
•
•
NIPS Workshop IRASL 2018
We visualize the outputs of such connections, projected back to the spectral domain, in models trained for speech denoising, and show that while skip connections do not necessarily improve performance with regards to the number of parameters, they make speech enhancement models more interpretable.
Speech Denoising by Accumulating Per-Frequency Modeling Fluctuations
mosheman5/DNP
•
•
The method is completely unsupervised and only trains on the specific audio clip that is being denoised.
Boosted Locality Sensitive Hashing: Discriminative Binary Codes for Source Separation
Speech enhancement tasks have seen significant improvements with the advance of deep learning technology, but with the cost of increased computational complexity.
