SepMamba：使用 Mamba 进行说话人分离的状态空间模型
SepMamba: State-space models for speaker separation using Mamba
AI总结：
本文提出主要由双向 Mamba 层构成的 U-Net 架构 SepMamba，用于单通道说话人分离，并在 WSJ0 2-speaker 上以更低计算、内存和前向时间超过同规模主流模型。
摘要：
近年来，基于深度学习的单通道说话人分离性能显著提升，这主要得益于基于 Transformer 的注意力机制的引入。然而，这些提升以高昂的计算需求为代价，使其无法用于许多实际应用。作为具有相似建模能力且计算高效的替代方案，Mamba 最近被提出。我们提出 SepMamba，这是一种主要由双向 Mamba 层组成的基于 U-Net 的架构。我们发现，在 WSJ0 2-speaker 数据集上，我们的方法优于包括基于 Transformer 的模型在内的类似规模的知名模型，同时显著降低了计算成本、内存使用量和前向传播时间。我们还报告了 SepMamba 因果变体的出色结果。对于深度语音分离而言，我们的方法为基于 Transformer 的架构提供了一种在计算上更具优势的替代方案。
英文摘要：
Deep learning-based single-channel speaker separation has improved significantly in recent years largely due to the introduction of the transformer-based attention mechanism. However, these improvements come at the expense of intense computational demands, precluding their use in many practical applications. As a computationally efficient alternative with similar modeling capabilities, Mamba was recently introduced. We propose SepMamba, a U-Net-based architecture composed primarily of bidirectional Mamba layers. We find that our approach outperforms similarly-sized prominent models - including transformer-based models - on the WSJ0 2-speaker dataset while enjoying a significant reduction in computational cost, memory usage, and forward pass time. We additionally report strong results for causal variants of SepMamba. Our approach provides a computationally favorable alternative to transformer-based architectures for deep speech separation.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
