Skipconvnet: Skip Convolutional Neural Network For Speech Dereverberation Using Optimally Smoothed Spectral Mapping

Abstract

The reliability of using fully convolutional networks (FCNs) has been successfully demonstrated by recent studies in many speech applications. One of the most popular variants of these FCNs is the `U-Net', which is an encoder-decoder network with skip connections. In this study, we propose `SkipConvNet' where we replace each skip connection with multiple convolutional modules to provide decoder with intuitive feature maps rather than encoder's output to improve the learning capacity of the network. We also propose the use of optimal smoothing of power spectral density (PSD) as a pre-processing step, which helps to further enhance the efficiency of the network. To evaluate our proposed system, we use the REVERB challenge corpus to assess the performance of various enhancement approaches under the same conditions. We focus solely on monitoring improvements in speech quality and their contribution to improving the efficiency of back-end speech systems, such as speech recognition and speak

Stats

• citations22
• S2 citations—
• github stars0
• HF likes0
• heat score10.21
• arxiv keykothapally2020skipconvnet

Related papers

• Speech Dereverberation Using Fully Convolutional Networks (2018)13.34
• Skipconvgan: Monaural Speech Dereverberation Using Generative Adversarial Networks Via Complex Time-frequency Masking (2022)9.92
• Deep Convolutional Neural Network-based Inverse Filtering Approach For Speech De-reverberation (2020)7.16
• Towards Speech Enhancement Using A Variational U-net Architecture (2020)7.81
• Inference Skipping For More Efficient Real-time Speech Enhancement With Parallel Rnns (2022)10.35
• Spatialnet: Extensively Learning Spatial Information For Multichannel Joint Speech Separation, Denoising And Dereverberation (2023)13.88
• Using Recurrences In Time And Frequency Within U-net Architecture For Speech Enhancement (2018)8.35
• Complex Spectral Mapping With Attention Based Convolution Recurrent Neural Network For Speech Enhancement (2021)0.00