BSDB-Net: Band-Split Dual-Branch Network with Selective State Spaces Mechanism for Monaural Speech Enhancement

2024-12-26

AI Summary

To address the compensation effect induced by amplitude-phase coupling and excessive model complexity in single-channel speech enhancement, this paper proposes a dual-branch network integrating band-splitting and Mamba-based state-space modeling. The method introduces a novel amplitude-phase decoupled dual-branch interaction architecture that enables independent optimization of magnitude and phase via complex spectral modeling. Band-splitting is employed to compress the frequency-domain dimensionality, while a lightweight Mamba-driven time-frequency joint modeling module is designed to achieve linear-complexity sequence modeling. Experiments demonstrate that the proposed approach reduces computational complexity by 8.3× on average over mainstream methods and by 25× compared to Transformer-based models, while maintaining state-of-the-art performance on PESQ, STOI, and other metrics—achieving significant improvements in both efficiency and generalization.

Technology Category

• Machine Learning: Mixture of Experts (MoE)
• Natural Language Processing: Speech
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Large language models for search
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Although the complex spectrum-based speech enhancement(SE) methods have achieved significant performance, coupling amplitude and phase can lead to a compensation effect, where amplitude information is sacrificed to compensate for the phase that is harmful to SE. In addition, to further improve the performance of SE, many modules are stacked onto SE, resulting in increased model complexity that limits the application of SE. To address these problems, we proposed a dual-path network based on compressed frequency using Mamba. First, we extract amplitude and phase information through parallel dual branches. This approach leverages structured complex spectra to implicitly capture phase information and solves the compensation effect by decoupling amplitude and phase, and the network incorporates an interaction module to suppress unnecessary parts and recover missing components from the other branch. Second, to reduce network complexity, the network introduces a band-split strategy to compress the frequency dimension. To further reduce complexity while maintaining good performance, we designed a Mamba-based module that models the time and frequency dimensions under linear complexity. Finally, compared to baselines, our model achieves an average 8.3 times reduction in computational complexity while maintaining superior performance. Furthermore, it achieves a 25 times reduction in complexity compared to transformer-based models.

Problem

• Monoaural Sound Clarification
• Size and Angle Interference
• System Complexity

Innovation

• Dual-Path Processing
• Mamba Technique
• Audio Band-Splitting