arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2406.09161cs.SDeess.AS
收藏

用于音频去噪的复数图像生成扩散Transformer

Complex Image-Generative Diffusion Transformer for Audio Denoising

Cornell University(康奈尔大学);Yeshiva University(叶史瓦大学)

Junhui Li, Pu Wang, Jialu Li, Youshan Zhang

AI总结:

本文提出复数图像生成扩散Transformer,将Transformer与扩散模型结合,利用复数傅里叶域信息和注意力扩散扩展感受野,在音频去噪任务上优于现有方法。

摘要:

音频去噪技术在深度神经网络领域引起了广泛关注。近年来,音频去噪问题已被转化为图像生成任务,基于深度学习的方法被应用于解决该问题。然而,其性能仍然有限,仍有进一步改进的空间。为了提高音频去噪性能,本文引入了一种复数图像生成扩散Transformer,该模型能够从复数傅里叶域中捕获更多信息。我们通过将Transformer与扩散模型相结合,探索了一种新颖的扩散Transformer。所提出的模型展示了Transformer的可扩展性,并利用注意力扩散扩展了稀疏注意力的感受野。我们的工作属于最早利用扩散Transformer处理音频去噪图像生成任务的研究之一。在两个基准数据集上的大量实验表明,我们提出的模型优于最先进的方法。

英文摘要:

The audio denoising technique has captured widespread attention in the deep neural network field. Recently, the audio denoising problem has been converted into an image generation task, and deep learning-based approaches have been applied to tackle this problem. However, its performance is still limited, leaving room for further improvement. In order to enhance audio denoising performance, this paper introduces a complex image-generative diffusion transformer that captures more information from the complex Fourier domain. We explore a novel diffusion transformer by integrating the transformer with a diffusion model. Our proposed model demonstrates the scalability of the transformer and expands the receptive field of sparse attention using attention diffusion. Our work is among the first to utilize diffusion transformers to deal with the image generation task for audio denoising. Extensive experiments on two benchmark datasets demonstrate that our proposed model outperforms state-of-the-art methods.

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。

↑