mlsanityData to 7 Oct 2026

Dual-path Mamba: Short and Long-term Bidirectional Selective Structured State Space Models for Speech Separation

arXiv 2403.18257 ↗First posted 27 Mar 2024eess.AS cs.SDcited by 45

On the map: State-space sequence models (S4, Mamba) → Speech and audio SSMs

Abstract (arXiv)

Transformers have been the most successful architecture for various speech modeling tasks, including speech separation. However, the self-attention mechanism in transformers with quadratic complexity is inefficient in computation and memory. Recent models incorporate new layers and modules along with transformers for better performance but also introduce extra model complexity. In this work, we replace transformers with Mamba, a selective state space model, for speech separation. We propose dual-path Mamba, which models short-term and long-term forward and backward dependency of speech signals using selective state spaces. Our experimental results on the WSJ0-2mix data show that our dual-path Mamba models of comparably smaller sizes outperform state-of-the-art RNN model DPRNN, CNN model WaveSplit, and transformer model Sepformer. Code: https://github.com/xi-j/Mamba-TasNet

Key facts. Cited by 45 papers in the map (data through 7 Oct 2026). Placed in Speech and audio SSMs, one of 10 directions in State-space sequence models (S4, Mamba). Builds on 32 of its 42 references that are in the map. First posted on arXiv 27 Mar 2024.

Builds on

32 of 42 references.

References · 32

  1. 2017Attention Is All You Needcited by 59,932
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2023Mamba: Linear-Time Sequence Modeling with Selective State Spacescited by 4,616
  4. 2016Gaussian Error Linear Units (GELUs)cited by 3,431
  5. 2021Efficiently Modeling Long Sequences with Structured State Spacescited by 2,424
  6. 2019Root Mean Square Layer Normalizationcited by 1,425
  7. 2024Vision Mamba: Efficient Visual Representation Learning with Bidirectional State Space Modelcited by 1,051
  8. 2021Combining Recurrent, Convolutional, and Continuous-time Models with Linear State-Space Layerscited by 607
  9. 2018SDR - half-baked or well done?cited by 420
  10. 2024U-Mamba: Enhancing Long-range Dependency for Biomedical Image Segmentationcited by 412
  11. 2021SpeechBrain: A General-Purpose Speech Toolkitcited by 404
  12. 2018Conv-TasNet: Surpassing Ideal Time-Frequency Magnitude Masking for Speech Separationcited by 400
20 more
  1. 2015Deep clustering: Discriminative embeddings for segmentation and separationcited by 211
  2. 2020Attention is All You Need in Speech Separationcited by 184
  3. 2019Dual-path RNN: efficient long sequence modeling for time-domain single-channel speech separationcited by 182
  4. 2024PointMamba: A Simple State Space Model for Point Cloud Analysiscited by 166
  5. 2024Graph-Mamba: Towards Long-Range Graph Sequence Modeling with Selective State Spacescited by 132
  6. 2017TasNet: time-domain audio separation network for real-time, single-channel speech separationcited by 115
  7. 2024Motion Mamba: Efficient and Long Sequence Motion Generationcited by 111
  8. 2020Wavesplit: End-to-End Speech Separation by Speaker Clusteringcited by 74
  9. 2020Dual-Path Transformer Network: Direct Context-Aware Modeling for End-to-End Monaural Speech Separationcited by 67
  10. 2020Sudo rm -rf: Efficient Networks for Universal Audio Source Separationcited by 50
  11. 2020Voice Separation with an Unknown Number of Multiple Speakerscited by 46
  12. 2023MossFormer2: Combining Transformer and RNN-Free Recurrent Network for Enhanced Time-Domain Monaural Speech Separationcited by 39
  13. 2024Multichannel Long-Term Streaming Neural Speech Enhancement for Static and Moving Speakerscited by 28
  14. 2023MossFormer: Pushing the Performance Limit of Monaural Speech Separation using Gated Single-Head Transformer with Convolution-Augmented Joint Self-Attentionscited by 27
  15. 2016Quantifying the vanishing gradient and long distance dependency problem in recursive neural networks and recursive LSTMscited by 21
  16. 2021Compute and memory efficient universal sound source separationcited by 15
  17. 2017End-to-end Source Separation with Adaptive Front-Endscited by 6
  18. 2023A Neural State-Space Model Approach to Efficient Speech Separationcited by 6
  19. 2022Exploring Self-Attention Mechanisms for Speech Separationcited by 4
  20. 2023On Time Domain Conformer Models for Monaural Speech Separation in Noisy Reverberant Acoustic Environmentscited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 24; 2025: 14; 2026: 7.

45 papers.

Sort:

By area:

  1. 2024DiM: Diffusion Mamba for Efficient High-Resolution Image Synthesiscited by 72
  2. 2024Mamba-360: Survey of State Space Models as Transformer Alternative for Long Sequence Modelling: Methods, Applications, and Challengescited by 63
  3. 2024An Investigation of Incorporating Mamba for Speech Enhancementcited by 61
  4. 2024State Space Model for New-Generation Network Alternative to Transformers: A Surveycited by 59
  5. 2024Mamba in Speech: Towards an Alternative to Self-Attentioncited by 58
  6. 2024MambaMixer: Efficient Selective State Space Models with Dual Token and Channel Selectioncited by 40
  7. 2024A Survey of Mambacited by 39
  8. 2024SSAMBA: Self-Supervised Audio Representation Learning with Mamba State Space Modelcited by 36
  9. 2025SepPrune: Structured Pruning for Efficient Deep Speech Separationcited by 31
  10. 2024Speech Slytherin: Examining the Performance and Efficiency of Mamba for Speech Separation, Recognition, and Synthesiscited by 22
  11. 2024SELD-Mamba: Selective State-Space Model for Sound Event Localization and Detection with Source Distance Estimationcited by 6
  12. 2024SAV-SE: Scene-aware Audio-Visual Speech Enhancement with Selective State Space Modelcited by 6
33 more
  1. 2025Fake-Mamba: Real-Time Speech Deepfake Detection Using Bidirectional Mamba as Self-Attention's Alternativecited by 6
  2. 2024TF-Mamba: A Time-Frequency Network for Sound Source Localizationcited by 5
  3. 2025Towards Efficient and Multifaceted Computer-assisted Pronunciation Training Leveraging Hierarchical Selective State Space Model and Decoupled Cross-entropy Losscited by 5
  4. 2024HeteGraph-Mamba: Heterogeneous Graph Learning via Selective State Space Modelcited by 3
  5. 2024Learning 1D Causal Visual Representation with De-focus Attention Networkscited by 3
  6. 2024DeFT-Mamba: Universal Multichannel Sound Separation and Polyphonic Audio Classificationcited by 3
  7. 2024Mamba-based Segmentation Model for Speaker Diarizationcited by 3
  8. 2025AAD-LLM: Neural Attention-Driven Auditory Scene Understandingcited by 3
  9. 2025Surgical-MambaLLM: Mamba2-enhanced Multimodal Large Language Model for VQLA in Robotic Surgerycited by 3
  10. 2024Joint Selective State Space Model and Detrending for Robust Time Series Anomaly Detectioncited by 2
  11. 2024CleanUMamba: A Compact Mamba Network for Speech Denoising using Channel Pruningcited by 2
  12. 2025Deep Active Speech Cancellation with Mamba-Masking Networkcited by 2
  13. 2025Technologies on Effectiveness and Efficiency: A Survey of State Spaces Modelscited by 2
  14. 2025An Exploration of Mamba for Speech Self-Supervised Modelscited by 2
  15. 2025Compression Method for Deep Diagonal State Space Model Based on H2 Optimal Reductioncited by 2
  16. 2026A Decomposition-based State Space Model for Multivariate Time-Series Forecastingcited by 2
  17. 2024Venturing into Uncharted Waters: The Navigation Compass from Transformer to Mambacited by 1
  18. 2024EDCSSM: Edge Detection with Convolutional State Space Modelcited by 1
  19. 2024Leveraging Joint Spectral and Spatial Learning with MAMBA for Multichannel Speech Enhancementcited by 1
  20. 2024Joint Fine-tuning and Conversion of Pretrained Speech and Language Models towards Linear Complexitycited by 1
  21. 2024ShadowMamba: State-Space Model with Boundary-Region Selective Scan for Shadow Removalcited by 1
  22. 2024Selective State Space Model for Monaural Speech Enhancementcited by 1
  23. 2025MLMA: Towards Multilingual ASR With Mamba-based Architecturescited by 1
  24. 2026Omni-directional attention mechanism based on Mamba for speech separationcited by 1
  25. 2025Active Speech Enhancement: Active Speech Denoising Decliping and Deveraberationcited by 0
  26. 2025JCAPT: A Joint Modeling Approach for CAPTcited by 0
  27. 2025Keyword Mamba: Spoken Keyword Spotting with State Space Modelscited by 0
  28. 2025Mamba2 Meets Silence: Robust Vocal Source Separation for Sparse Regionscited by 0
  29. 2026A Mamba-Based Model for Automatic Chord Recognitioncited by 0
  30. 2026TokenSE: a Mamba-based discrete token speech enhancement framework for cochlear implantscited by 0
  31. 2026Query-based Cross-Modal Projector Bolstering Mamba Multimodal LLMcited by 0
  32. 2026MambAdapter: Lightweight Mamba-Based Adapters for Parameter-Efficient Transfer Learning in Speech and Audiocited by 0
  33. 2026State Space Models Meet Remote Sensing: A Surveycited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024SepMamba: State-space models for speaker separation using Mamba
  2. 2024Speech-Mamba: Long-Context Speech Recognition with Selective State Spaces Models
  3. 2024SPMamba: State-space model is all you need in speech separation
  4. 2024U-Mamba-Net: A highly efficient Mamba-based U-net style network for noisy and reverberant speech separation
  5. 2024Transformers are SSMs: Generalized Models and Efficient Algorithms Through Structured State Space Duality
  6. 2024Mamba-based Decoder-Only Approach with Bidirectional Speech Modeling for Speech Recognition
  7. 2024Exploring the Capability of Mamba in Speech Applications
  8. 2024An Empirical Study of Mamba-based Language Models