mlsanityData to 7 Oct 2026

TSTNN: Two-stage Transformer based Neural Network for Speech Enhancement in the Time Domain

arXiv 2103.09963 ↗First posted 18 Mar 2021eess.AS cs.LG cs.SDcited by 36

Not on the map. Why?

Abstract (arXiv)

In this paper, we propose a transformer-based architecture, called two-stage transformer neural network (TSTNN) for end-to-end speech denoising in the time domain. The proposed model is composed of an encoder, a two-stage transformer module (TSTM), a masking module and a decoder. The encoder maps input noisy speech into feature representation. The TSTM exploits four stacked two-stage transformer blocks to efficiently extract local and global information from the encoder output stage by stage. The masking module creates a mask which will be multiplied with the encoder output. Finally, the decoder uses the masked encoder feature to reconstruct the enhanced speech. Experimental results on the benchmark dataset show that the TSTNN outperforms most state-of-the-art models in time or frequency domain while having significantly lower model complexity.

Key facts. Cited by 36 papers in the map (data through 7 Oct 2026). Builds on 16 of its 28 references that are in the map. First posted on arXiv 18 Mar 2021.

Builds on

16 of 28 references.

References · 16

  1. 2017Attention Is All You Needcited by 59,932
  2. 2016Layer Normalizationcited by 5,579
  3. 2015Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classificationcited by 3,590
  4. 2016Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Networkcited by 1,199
  5. 2015Multi-Scale Context Aggregation by Dilated Convolutionscited by 1,128
  6. 2020Real Time Speech Enhancement in the Waveform Domaincited by 183
  7. 2019Dual-path RNN: efficient long sequence modeling for time-domain single-channel speech separationcited by 182
  8. 2017SEGAN: Speech Enhancement Generative Adversarial Networkcited by 165
  9. 2019MetricGAN: Generative Adversarial Networks based Black-box Metric Scores Optimization for Speech Enhancementcited by 104
  10. 2019Phase-aware Speech Enhancement with Deep Complex U-Netcited by 73
  11. 2020Dual-Path Transformer Network: Direct Context-Aware Modeling for End-to-End Monaural Speech Separationcited by 67
  12. 2017A Wavenet for Speech Denoisingcited by 47
4 more
  1. 2019T-GSA: Transformer with Gaussian-weighted self-attention for speech enhancementcited by 34
  2. 2018Self-Attentional Acoustic Modelscited by 29
  3. 2018Improved Speech Enhancement with the Wave-U-Netcited by 23
  4. 2017Complex spectrogram enhancement by convolutional neural network with multi-metrics learningcited by 17

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 2; 2022: 6; 2023: 8; 2024: 9; 2025: 8; 2026: 3.

36 papers.

Sort:

By area:

  1. 2024Automatic Speech Recognition using Advanced Deep Learning Approaches: A surveycited by 50
  2. 2022CMGAN: Conformer-based Metric GAN for Speech Enhancementcited by 45
  3. 2021Deep Spoken Keyword Spotting: An Overviewcited by 43
  4. 2023Transformers in Speech Processing: A Surveycited by 38
  5. 2022CMGAN: Conformer-Based Metric-GAN for Monaural Speech Enhancementcited by 22
  6. 2021Dual-branch Attention-In-Attention Transformer for single-channel speech enhancementcited by 18
  7. 2024Mamba-SEUNet: Mamba UNet for Monaural Speech Enhancementcited by 10
  8. 2024BSS-CFFMA: Cross-Domain Feature Fusion and Multi-Attention Speech Enhancement Network based on Self-Supervised Embeddingcited by 9
  9. 2022BirdSoundsDenoising: Deep Visual Audio Denoising for Bird Soundscited by 7
  10. 2022Multi-View Attention Transfer for Efficient Speech Enhancementcited by 6
  11. 2024TrustEMG-Net: Using Representation-Masking Transformer with U-Net for Surface Electromyography Enhancementcited by 6
  12. 2024MUSE: Flexible Voiceprint Receptive Fields and Multi-Path Fusion Enhanced Taylor Transformer for U-Net-based Speech Enhancementcited by 4
24 more
  1. 2022Deep neural network techniques for monaural speech enhancement: state of the art analysiscited by 3
  2. 2023Complex Image Generation SwinTransformer Network for Audio Denoisingcited by 3
  3. 2024DeFT-Mamba: Universal Multichannel Sound Separation and Polyphonic Audio Classificationcited by 3
  4. 2024DENOASR: Debiasing ASRs through Selective Denoisingcited by 3
  5. 2025Linguistic Knowledge Transfer Learning for Speech Enhancementcited by 3
  6. 2023Efficient Encoder-Decoder and Dual-Path Conformer for Comprehensive Feature Learning in Speech Enhancementcited by 2
  7. 2023DPATD: Dual-Phase Audio Transformer for Denoisingcited by 2
  8. 2023Physics-Informed Data Denoising for Real-Life Sensing Systemscited by 2
  9. 2023ROSE: A Recognition-Oriented Speech Enhancement Framework in Air Traffic Control Using Multi-Objective Learningcited by 2
  10. 2022Audio Denoising for Robust Audio Fingerprintingcited by 1
  11. 2024Diffusion Gaussian Mixture Audio Denoisecited by 1
  12. 2025Throat and acoustic paired speech dataset for deep learning-based speech enhancementcited by 1
  13. 2025HiFi-Stream: Streaming Speech Enhancement with Generative Adversarial Networkscited by 1
  14. 2025A Low-Power Streaming Speech Enhancement Accelerator For Edge Devicescited by 1
  15. 2026Quaternion Self-Attention with Shared Scorescited by 1
  16. 2023Music Augmentation and Denoising For Peak-Based Audio Fingerprintingcited by 0
  17. 2023Improving Label Assignments Learning by Dynamic Sample Dropout Combined with Layer-wise Optimization in Speech Separationcited by 0
  18. 2024Survey: Transformer-based Models in Data Modality Conversioncited by 0
  19. 2025UL-UNAS: Ultra-Lightweight U-Nets for Real-Time Speech Enhancement via Network Architecture Searchcited by 0
  20. 2025Active Speech Enhancement: Active Speech Denoising Decliping and Deveraberationcited by 0
  21. 2025Modality-Specific Speech Enhancement and Noise-Adaptive Fusion for Acoustic and Body-Conduction Microphone Frameworkcited by 0
  22. 2025IMSE: Efficient U-Net-based Speech Enhancement using Inception Depthwise Convolution and Amplitude-Aware Linear Attentioncited by 0
  23. 2026DBHN-Net: Dual-Branch Hybrid Neural Network For Low-Complexity Monaural Speech Enhancementcited by 0
  24. 2026Neuromorphic Speech Enhancement with Dual-Branch Spiking Neural Networkscited by 0