Future of Information and Communication Conference (FICC) 2023
2-3 March 2023
Publication Links
IJACSA
Special Issues
Article Details
Copyright Statement: This is an open access article licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution, and reproduction in any medium, even commercially as long as the original work is properly cited.
Dilated Multi-Activation Autoencoder to Improve the Performance of Sound Separation Mechanisms
Digital Object Identifier (DOI) : 10.14569/IJACSA.2022.0131231
Article Published in International Journal of Advanced Computer Science and Applications(IJACSA), Volume 13 Issue 12, 2022.
Abstract: Speech enhancement is the process of improving the quality of audio relative to target speaker while suppressing other sounds. It can be used in many applications as speech recognition, mobile phone, hearing aids and also enhancing audio files resulted from separation models. In this paper, a convolutional neural network (CNN) architecture is proposed to improve the quality of target’s speaker resulted from speech separation models without having any prior information about the background sounds. The proposed model consists of three main phases: Pre-Processing phase, Autoencoder phase and Retrieving Audio phase. The pre-processing phase converts audio to short time Fourier transform (STFT) domain. Autoencoder phase consists of two main modules: dilated multi-Activation encoder and dilated multi-Activation decoder. Dilated multi-Activation encoder module has a six blocks with different dilation factors and each block consists of three CNN layers where each layer has different activation function then the encoder’s blocks are arranged in reverse order to construct dilated multi-activation decoder. Audio retrieving phase is used to reconstruct audio depending on feature resulted from second phase. Audio files resulted from separation models are used to build our datasets that consist of 31250 files. The proposed dilated multi-activation autoencoder improved separated audios Segmental Signal-to-Noise Ratio (SNRseg) with 33.9%, Short-time objective intelligibility (STOI) with 1.3% and reduced bark spectral distortion (BSD) with 97%.
Keywords: Speech de-noising; speech enhancement; speech separation; short time Fourier transform (STFT); autoencoder; dilated Convolution neural network; multi-activation functions; convolution neural network (CNN); bidirectional long short memory (BLSTM)
Ghada Dahy, Mohammed A. A. Refaey, Reda Alkhoribi and M. Shoman, “Dilated Multi-Activation Autoencoder to Improve the Performance of Sound Separation Mechanisms” International Journal of Advanced Computer Science and Applications(IJACSA), 13(12), 2022. http://dx.doi.org/10.14569/IJACSA.2022.0131231
@article{Dahy2022,
title = {Dilated Multi-Activation Autoencoder to Improve the Performance of Sound Separation Mechanisms},
journal = {International Journal of Advanced Computer Science and Applications},
doi = {10.14569/IJACSA.2022.0131231},
url = {http://dx.doi.org/10.14569/IJACSA.2022.0131231},
year = {2022},
publisher = {The Science and Information Organization},
volume = {13},
number = {12},
author = {Ghada Dahy and Mohammed A. A. Refaey and Reda Alkhoribi and M. Shoman}
}
