统一麦克风转换:通过特征级线性调制实现多对多设备映射
Unified Microphone Conversion: Many-to-Many Device Mapping via Feature-wise Linear Modulation
AI总结:
针对声音事件分类系统受设备差异影响、现有转换方法可扩展性差的问题,提出基于频率响应条件和特征级线性调制的统一生成框架,实现多对多设备映射,性能更优。
摘要:
我们提出了统一麦克风转换(Unified Microphone Conversion),这是一个统一的生成式框架,旨在增强声音事件分类(SEC)系统抵御设备差异影响的能力。我们此前基于CycleGAN的方法虽能有效模拟设备特性,但每对设备都需要单独的模型,限制了可扩展性。我们的方法通过让生成器以频率响应数据为条件,经非配对训练实现多对多设备映射,突破了这一限制。我们通过特征级线性调制(Feature-wise Linear Modulation)整合频率响应信息,进一步提升了可扩展性。此外,引入合成的频率响应差异提升了该框架在实际场景中的适用性。实验结果表明,我们的方法在宏平均F1值上优于当前最优方法2.6%,并将差异降低了0.8%。
英文摘要:
We present Unified Microphone Conversion, a unified generative framework designed to bolster sound event classification (SEC) systems against device variability. While our prior CycleGAN-based methods effectively simulate device characteristics, they require separate models for each device pair, limiting scalability. Our approach overcomes this constraint by conditioning the generator on frequency response data, enabling many-to-many device mappings through unpaired training. We integrate frequency-response information via Feature-wise Linear Modulation, further enhancing scalability. Additionally, incorporating synthetic frequency response differences improves the applicability of our framework for real-world application. Experimental results show that our method outperforms the state-of-the-art by 2.6% and reduces variability by 0.8% in macro-average F1 score.
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。