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

Abstract

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

Stats

• citations0
• S2 citations—
• github stars114
• HF likes0
• heat score4.12
• arxiv keyjiang2024dual

Code

• xi-j/Mamba-TasNet—

Related papers

• Mamba-based Decoder-only Approach With Bidirectional Speech Modeling For Speech Recognition (2024)0.00
• Mamba-seunet: Mamba Unet For Monaural Speech Enhancement (2024)7.16
• Audio Mamba: Bidirectional State Space Model For Audio Representation Learning (2024)11.58
• Speech Slytherin: Examining The Performance And Efficiency Of Mamba For Speech Separation, Recognition, And Synthesis (2024)13.88
• Audio Mamba: Selective State Spaces For Self-supervised Audio Representations (2024)9.23
• U-mamba-net: A Highly Efficient Mamba-based U-net Style Network For Noisy And Reverberant Speech Separation (2024)4.52
• Mamba2 Meets Silence: Robust Vocal Source Separation For Sparse Regions (2025)0.00
• An Exploration Of Mamba For Speech Self-supervised Models (2025)1.20