Speech Enhancement

Speech Enhancement is a signal processing task that involves improving the quality of speech signals captured under noisy or degraded conditions. The goal of speech enhancement is to make speech signals clearer, more intelligible, and more pleasant to listen to, which can be used for various applications such as voice recognition, teleconferencing, and hearing aids. A representative Github project with online demo : ClearerVoice-Studio.

( Image credit: A Fully Convolutional Neural Network For Speech Enhancement )

Papers

Recently Added Most Hyped Most Active Needs Verification Most Verified

Showing 476–500 of 982 papers

Title	Date	Tasks	Status	Hype
NeuralEcho: A Self-Attentive Recurrent Neural Network For Unified Acoustic Echo Suppression And Speech Enhancement	May 20, 2022	Acoustic echo cancellationSpeech Enhancement	—Unverified	0
U-Former: Improving Monaural Speech Enhancement with Multi-head Self and Cross Attention	May 18, 2022	DecoderSpeech Enhancement	CodeCode Available	1
Dictionary-Based Fusion of Contact and Acoustic Microphones for Wind Noise Reduction	May 18, 2022	Speech Enhancement	—Unverified	0
Streaming Noise Context Aware Enhancement For Automatic Speech Recognition in Multi-Talker Environments	May 17, 2022	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	—Unverified	0
Task splitting for DNN-based acoustic echo and noise removal	May 13, 2022	Acoustic echo cancellationSpeech Enhancement	—Unverified	0
DeepFilterNet2: Towards Real-Time Speech Enhancement on Embedded Devices for Full-Band Audio	May 11, 2022	CPUData Augmentation	CodeCode Available	4
A deep representation learning speech enhancement method using β-VAE	May 11, 2022	DisentanglementRepresentation Learning	—Unverified	0
Real-Time Packet Loss Concealment With Mixed Generative and Predictive Model	May 11, 2022	Packet Loss ConcealmentSpeech Enhancement	CodeCode Available	3
Generalized Fast Multichannel Nonnegative Matrix Factorization Based on Gaussian Scale Mixtures for Blind Source Separation	May 11, 2022	blind source separationSpeech Enhancement	—Unverified	0
Speaker Reinforcement Using Target Source Extraction for Robust Automatic Speech Recognition	May 9, 2022	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	—Unverified	0
Acoustic echo suppression using a learning-based multi-frame minimum variance distortionless response filter	May 7, 2022	parameter estimationSpeech Enhancement	—Unverified	0
On monoaural speech enhancement for automatic recognition of real noisy speech using mixture invariant training	May 3, 2022	Robust Speech RecognitionSpeech Enhancement	—Unverified	0
Improving Dual-Microphone Speech Enhancement by Learning Cross-Channel Features with Multi-Head Attention	May 3, 2022	DecoderMulti-Task Learning	—Unverified	0
A Meeting Transcription System for an Ad-Hoc Acoustic Sensor Network	May 2, 2022	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	—Unverified	0
Improved far-field speech recognition using Joint Variational Autoencoder	Apr 24, 2022	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	—Unverified	0
RadioSES: mmWave-Based Audioradio Speech Enhancement and Separation System	Apr 14, 2022	Speech EnhancementSpeech Separation	—Unverified	0
Receptive Field Analysis of Temporal Convolutional Networks for Monaural Speech Dereverberation	Apr 13, 2022	Speech DereverberationSpeech Enhancement	CodeCode Available	0
VoiceFixer: A Unified Framework for High-Fidelity Speech Restoration	Apr 12, 2022	Speech DenoisingSpeech Enhancement	CodeCode Available	3
Listen only to me! How well can target speech extraction handle false alarms?	Apr 11, 2022	Speaker IdentificationSpeaker Verification	—Unverified	0
Exploiting Hidden Representations from a DNN-based Speech Recogniser for Speech Intelligibility Prediction in Hearing-impaired Listeners	Apr 8, 2022	PredictionSpeech Enhancement	CodeCode Available	0
Boosting Self-Supervised Embeddings for Speech Enhancement	Apr 7, 2022	Self-Supervised LearningSpeech Enhancement	CodeCode Available	1
FFC-SE: Fast Fourier Convolution for Speech Enhancement	Apr 6, 2022	Speech Enhancement	—Unverified	0
Expression-preserving face frontalization improves visually assisted speech processing	Apr 6, 2022	Face ModelLip Reading	—Unverified	0
Audio-visual multi-channel speech separation, dereverberation and recognition	Apr 5, 2022	Automatic Speech RecognitionAutomatic Speech Recognition (ASR)	—Unverified	0
Complex Recurrent Variational Autoencoder with Application to Speech Enhancement	Apr 5, 2022	Speech Enhancement	CodeCode Available	0

Show:10 25 50

← PrevPage 20 of 40Next →

All datasets VoiceBank + DEMAND Deep Noise Suppression (DNS) Challenge CHiME-3 EARS-WHAM EasyCom DNS Challenge VB-DemandEx WHAMR!WSJ0 + DEMAND + RNNoise RealMAN VoiceBank+DEMAND DEMAND

Benchmark Results

#	Model	Metric	Claimed	Verified	Status
1	ROSE-CD(PESQ)	PESQ (wb)	3.99	—	Unverified
2	PESQetarian	PESQ (wb)	3.82	—	Unverified
3	Mamba-SEUNet L (+PCS)	PESQ (wb)	3.73	—	Unverified
4	Schrödinger bridge (PESQ loss)	PESQ (wb)	3.7	—	Unverified
5	SEMamba (+PCS)	PESQ (wb)	3.69	—	Unverified
6	ZipEnhancer (S, \lamba_6 = 0)	PESQ (wb)	3.63	—	Unverified
7	PrimeK-Net	PESQ (wb)	3.61	—	Unverified
8	ZipEnhancer (S, \lamba_6 = 0.2)	PESQ (wb)	3.61	—	Unverified
9	MP-SENet	PESQ (wb)	3.6	—	Unverified
10	PCS_CS_WAVLM	PESQ (wb)	3.54	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	BSRNN-S + MGD	SI-SDR-WB	21.4	—	Unverified
2	DTLN	SI-SDR-WB	16.34	—	Unverified
3	Non-Real-Time MultiScale+	SI-SDR-WB	16.22	—	Unverified
4	ZipEnhancer (M)	PESQ-WB	3.81	—	Unverified
5	TF-Locoformer (M)	PESQ-WB	3.72	—	Unverified
6	ZipEnhancer (S)	PESQ-WB	3.69	—	Unverified
7	MambAttention	PESQ-WB	3.67	—	Unverified
8	MP-SENet	PESQ-WB	3.62	—	Unverified
9	xLSTM-SENet	PESQ-WB	3.59	—	Unverified
10	BSRNN-S + MRSD	PESQ-WB	3.53	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Inter-Channel Conv-TasNet	SDR	19.67	—	Unverified
2	CA Dense U-Net (Complex)	SDR	18.64	—	Unverified
3	Dense U-Net (Complex)	SDR	18.4	—	Unverified
4	Dense U-Net (Real)	SDR	16.86	—	Unverified
5	U-Net (Real)	SDR	15.97	—	Unverified
6	Noisy/unprocessed	SDR	6.5	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Schrödinger Bridge (PESQ loss)	PESQ-WB	3.09	—	Unverified
2	SGMSE+	PESQ-WB	2.5	—	Unverified
3	Demucs v4	PESQ-WB	2.37	—	Unverified
4	Schrödinger Bridge	PESQ-WB	2.33	—	Unverified
5	Conv-TasNet	PESQ-WB	2.31	—	Unverified
6	CDiffuSE	PESQ-WB	1.6	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	ReVISE (ch2)	Audio Quality MOS	4.19	—	Unverified
2	ReVISE (bf)	Audio Quality MOS	4.11	—	Unverified
3	Demucs (ch2)	Audio Quality MOS	2.95	—	Unverified
4	Demucs (bf)	Audio Quality MOS	2.39	—	Unverified
5	MaxDI (Baseline)	PESQ	1.17	—	Unverified
6	DAJA (MVDR,HMA,1000) (Overlapped Speech)	SDR	-4.76	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	ZipEnhancer (M)	PESQ-NB	4.08	—	Unverified
2	DCCRN-MC	PESQ-NB	3.21	—	Unverified
3	DCCRN-M	PESQ-NB	3.15	—	Unverified
4	DCCRN	PESQ-NB	3.04	—	Unverified
5	RNN-Modulation	PESQ-WB	2.75	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	MambAttention	ESTOI	0.8	—	Unverified
2	SEMamba	ESTOI	0.8	—	Unverified
3	xLSTM-SENet	ESTOI	0.8	—	Unverified
4	MP-SENet	ESTOI	0.79	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SepFormer	PESQ	2.84	—	Unverified
2	DTLN	PESQ	2.23	—	Unverified
3	Unprocessed	PESQ	1.83	—	Unverified
4	Non-Real-Time MultiScale+	PESQ	1.52	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DCUNet-MC	PESQ-NB	3.44	—	Unverified
2	DCCRN-M	PESQ-NB	3.28	—	Unverified
3	DCUNet	PESQ-NB	3.25	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	CleanMel-L-map	DNSMOS	3.82	—	Unverified
2	SpatialNet	DNSMOS BAK	3.43	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	rose_cd(PESQ )	PESQ	3.99	—	Unverified
2	ROSE-CD	PESQ	3.49	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Wave-U-Net	CBAK	3.24	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Audio-Visual concat-ref	PESQ	2.7	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SE-MelGAN	Audio Quality MOS	3.1	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	DeFT-AN	PESQ	3.01	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	Audio-Visual concat-ref	PESQ	3.03	—	Unverified

#	Model	Metric	Claimed	Verified	Status
1	SepFormer	PESQ	3.07	—	Unverified