TridentSE：使用32个全局令牌指导语音增强
TridentSE: Guiding Speech Enhancement with 32 Global Tokens
摘要：
在本文中，我们提出了TridentSE，一种新的架构语音增强，这是能够有效地捕捉全局信息和局部细节。TridentSE维护T-F bin级别的表示来捕获细节，并使用少量的全局令牌来处理全局信息。信息通过交叉注意模块在局部和全局表示之间传播。为了捕获帧间和帧内信息，将全局令牌分为两组，分别沿时间轴和频率轴进行处理。一个度量指标，以指导我们的模型，以实现更高的感知质量。即使计算成本显著降低，TridentSE的性能也优于以前的各种语音增强方法，在VoiceBank+DEMAND数据集上实现了3.47的PESQ，在DNS无混响测试集上实现了3.44的PESQ。可视化显示，全局令牌学习不同的和可解释的全局模式。
英文摘要：
In this paper, we present TridentSE, a novel architecture for speech enhancement, which is capable of efficiently capturing both global information and local details. TridentSE maintains T-F bin level representation to capture details, and uses a small number of global tokens to process the global information. Information is propagated between the local and the global representations through cross attention modules. To capture both inter- and intra-frame information, the global tokens are divided into two groups to process along the time and the frequency axis respectively. A metric discriminator is further employed to guide our model to achieve higher perceptual quality. Even with significantly lower computational cost, TridentSE outperforms a variety of previous speech enhancement methods, achieving a PESQ of 3.47 on VoiceBank+DEMAND dataset and a PESQ of 3.44 on DNS no-reverb test set. Visualization shows that the global tokens learn diverse and interpretable global patterns.
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。
