Audio Denoising
Audio denoising aims to remove unwanted noise from sound recordings, improving audio quality and enabling more accurate analysis. Current research heavily utilizes deep learning, employing architectures like transformers, diffusion models, and autoencoders, often framing the problem as an image generation or segmentation task to leverage advanced visual processing techniques. This field is crucial for applications ranging from enhancing animal vocalizations for biodiversity studies to improving speech recognition and robust audio fingerprinting in music identification services, impacting various scientific disciplines and technological advancements.
Papers
Vision Transformer Segmentation for Visual Bird Sound Denoising
Sahil Kumar, Jialu Li, Youshan Zhang
Complex Image-Generative Diffusion Transformer for Audio Denoising
Junhui Li, Pu Wang, Jialu Li, Youshan Zhang
Diffusion Gaussian Mixture Audio Denoise
Pu Wang, Junhui Li, Jialu Li, Liangdong Guo, Youshan Zhang