arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2406.09154cs.SDcs.CLeess.AS
收藏

扩散高斯混合音频去噪

Diffusion Gaussian Mixture Audio Denoise

University of Science and Technology Liaoning(辽宁科技大学);Cornell University(康奈尔大学);Yeshiva University(叶史瓦大学)

Pu Wang, Junhui Li, Jialu Li, Liangdong Guo, Youshan Zhang

AI总结:

本文提出 DiffGMM 模型,结合扩散模型与高斯混合模型,通过 1D-U-Net 估计真实噪声分布并逐步去噪,在音频去噪任务中达到最先进性能。

摘要:

近年来,扩散模型在音频去噪任务中取得了令人瞩目的性能。其反向过程的独特性质能够恢复干净信号。然而,现实世界噪声的分布并不符合单一高斯分布,甚至可能是未知的。高斯噪声条件的采样限制了其应用场景。为克服这些挑战,我们提出了一种 DiffGMM 模型,这是一种基于扩散模型和高斯混合模型的去噪模型。我们利用反向过程来估计高斯混合模型的参数。给定一个带噪音频信号,我们首先应用 1D-U-Net 提取特征,并训练线性层来估计高斯混合模型的参数,从而近似真实噪声分布。带噪信号不断减去估计的噪声,以输出干净的音频信号。大量实验结果表明,所提出的 DiffGMM 模型达到了最先进的性能。

英文摘要:

Recent diffusion models have achieved promising performances in audio-denoising tasks. The unique property of the reverse process could recover clean signals. However, the distribution of real-world noises does not comply with a single Gaussian distribution and is even unknown. The sampling of Gaussian noise conditions limits its application scenarios. To overcome these challenges, we propose a DiffGMM model, a denoising model based on the diffusion and Gaussian mixture models. We employ the reverse process to estimate parameters for the Gaussian mixture model. Given a noisy audio signal, we first apply a 1D-U-Net to extract features and train linear layers to estimate parameters for the Gaussian mixture model, and we approximate the real noise distributions. The noisy signal is continuously subtracted from the estimated noise to output clean audio signals. Extensive experimental results demonstrate that the proposed DiffGMM model achieves state-of-the-art performance.

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。

↑