用十方向注意力让Mamba模型捕捉语音谱图全局依赖，提升分离效果。
Omni-directional attention mechanism based on Mamba for speech separation
- 在时频谱图上从十个方向建模全局依赖，突破传统单向处理限制。
- 在三个公开数据集上显著优于基线模型，性能超越当前最先进系统。
- 保持线性复杂度，适合长序列语音分离任务，兼顾效率与精度。
Mamba作为一种选择性状态空间模型（SSM），已成为语音建模中替代Transformer的高效方案，能以线性复杂度处理长序列。然而，现有方法无论在时域或时频域，通常沿单一维度将输入分解为短一维序列，再通过Mamba处理，这限制了其仅能进行局部一维建模，难以捕捉时频谱图上的全局依赖。本文提出一种基于单向Mamba的高效全向注意力（OA）机制，在谱图上从十个不同方向建模全局依赖。我们将该机制扩展至两个基线分离模型，并在三个公开数据集上评估。实验结果表明，所提方法在保持线性复杂度的同时，持续显著优于基线模型，性能超越现有最先进系统。
原文摘要 · Abstract (English)
Mamba, a selective state-space model (SSM), has emerged as an efficient alternative to Transformers for speech modeling, enabling long-sequence processing with linear complexity. While effective in speech separation, existing approaches, whether in the time or time-frequency domain, typically decompose the input along a single dimension into short one-dimensional sequences before processing them with Mamba, which restricts it to local 1D modeling and limits its ability to capture global dependencies across the 2D spectrogram. In this work, we propose an efficient omni-directional attention (OA) mechanism built upon unidirectional Mamba, which models global dependencies from ten different directions on the spectrogram. We expand the proposed mechanism into two baseline separation models and evaluate on three public datasets. Experimental results show that our approach consistently achieves significant performance gains over the baselines while preserving linear complexity, outperforming existing state-of-the-art (SOTA) systems.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。