Speech enhancement using deep neural network based on mask estimation and harmonic regeneration noise reduction for single channel microphone

The development of speech-enabled mobile applications has greatly improved human-computer interaction in recent years. These applications are flexible and convenient for users. Since the speech signal is captured in mobile conditions, it may easily be contaminated by background noises, which may...

Full description

Saved in:

Bibliographic Details
Main Author:	Md Jamal, Norezmi
Format:	Thesis
Language:	English English English
Published:	2022
Subjects:	TK5101-6720 Telecommunication Including telegraphy, telephone, radio, radar, television
Online Access:	http://eprints.uthm.edu.my/8464/1/24p%20NOREZMI%20MD%20JAMAL.pdf http://eprints.uthm.edu.my/8464/2/NOREZMI%20MD%20JAMAL%20COPYRIGHT%20DECLARATION.pdf http://eprints.uthm.edu.my/8464/3/NOREZMI%20MD%20JAMAL%20WATERMARK.pdf
Tags:	Add Tag No Tags, Be the first to tag this record!

id	my-uthm-ep.8464
record_format	uketd_dc
spelling	my-uthm-ep.84642023-02-27T01:01:10Z Speech enhancement using deep neural network based on mask estimation and harmonic regeneration noise reduction for single channel microphone 2022-01 Md Jamal, Norezmi TK5101-6720 Telecommunication. Including telegraphy, telephone, radio, radar, television The development of speech-enabled mobile applications has greatly improved human-computer interaction in recent years. These applications are flexible and convenient for users. Since the speech signal is captured in mobile conditions, it may easily be contaminated by background noises, which may result in a complicated computation and require speech enhancement algorithm. Thus, the performance of speech applications can be degraded when signal-to-noise ratio (SNR) is low and nonstationary noise is present. Moreover, the task of removing noises without causing speech distortion is also challenging, in which the quality and intelligibility of speech are affected. In order to overcome these issues, a supervised Deep Neural Network (DNN) algorithm predicted constrained Wiener Filter (cWF) target mask algorithm based on extracted Gammatone filter bank power spectrum (GF-TF) features and trained model is developed. As a result, the trained model with GF-TF features and cross-speech dataset produced promising results, while the proposed target mask scored higher on the perceptual evaluation of speech quality (PESQ) and short-time objective intelligibility (STOI) tests. On top of that, a modified Harmonic Regeneration Noise Reduction (HRNR) algorithm is proposed as a post-filtering strategy to enhance speech signal due to residual noise being introduced after DNN prediction. Results from TIMIT dataset revealed that average STOI scores for the joint algorithm are higher than those of DNN, conventional HRNR and Log Minimum Mean Square Error (Log-MMSE) algorithms. With SNR of -5 dB, an improvement of 4% over DNN algorithm, 36% over conventional HRNR algorithm, and 12% over Log-MMSE algorithm are obtained. While the average PESQ score is less affected after post-filtering strategy. Thus, this work has contributed to improve speech intelligibility from noisy backgrounds at low SNR as it can be deployed in speechenabled mobile applications. 2022-01 Thesis http://eprints.uthm.edu.my/8464/ http://eprints.uthm.edu.my/8464/1/24p%20NOREZMI%20MD%20JAMAL.pdf text en public http://eprints.uthm.edu.my/8464/2/NOREZMI%20MD%20JAMAL%20COPYRIGHT%20DECLARATION.pdf text en staffonly http://eprints.uthm.edu.my/8464/3/NOREZMI%20MD%20JAMAL%20WATERMARK.pdf text en validuser phd doctoral Universiti Tun Hussein Onn Malaysia Fakulti Kejuruteraan Elektrik dan Elektronik
institution	Universiti Tun Hussein Onn Malaysia
collection	UTHM Institutional Repository
language	English English English
topic	TK5101-6720 Telecommunication Including telegraphy, telephone, radio, radar, television
spellingShingle	TK5101-6720 Telecommunication Including telegraphy, telephone, radio, radar, television Md Jamal, Norezmi Speech enhancement using deep neural network based on mask estimation and harmonic regeneration noise reduction for single channel microphone
description	The development of speech-enabled mobile applications has greatly improved human-computer interaction in recent years. These applications are flexible and convenient for users. Since the speech signal is captured in mobile conditions, it may easily be contaminated by background noises, which may result in a complicated computation and require speech enhancement algorithm. Thus, the performance of speech applications can be degraded when signal-to-noise ratio (SNR) is low and nonstationary noise is present. Moreover, the task of removing noises without causing speech distortion is also challenging, in which the quality and intelligibility of speech are affected. In order to overcome these issues, a supervised Deep Neural Network (DNN) algorithm predicted constrained Wiener Filter (cWF) target mask algorithm based on extracted Gammatone filter bank power spectrum (GF-TF) features and trained model is developed. As a result, the trained model with GF-TF features and cross-speech dataset produced promising results, while the proposed target mask scored higher on the perceptual evaluation of speech quality (PESQ) and short-time objective intelligibility (STOI) tests. On top of that, a modified Harmonic Regeneration Noise Reduction (HRNR) algorithm is proposed as a post-filtering strategy to enhance speech signal due to residual noise being introduced after DNN prediction. Results from TIMIT dataset revealed that average STOI scores for the joint algorithm are higher than those of DNN, conventional HRNR and Log Minimum Mean Square Error (Log-MMSE) algorithms. With SNR of -5 dB, an improvement of 4% over DNN algorithm, 36% over conventional HRNR algorithm, and 12% over Log-MMSE algorithm are obtained. While the average PESQ score is less affected after post-filtering strategy. Thus, this work has contributed to improve speech intelligibility from noisy backgrounds at low SNR as it can be deployed in speechenabled mobile applications.
format	Thesis
qualification_name	Doctor of Philosophy (PhD.)
qualification_level	Doctorate
author	Md Jamal, Norezmi
author_facet	Md Jamal, Norezmi
author_sort	Md Jamal, Norezmi
title	Speech enhancement using deep neural network based on mask estimation and harmonic regeneration noise reduction for single channel microphone
title_short	Speech enhancement using deep neural network based on mask estimation and harmonic regeneration noise reduction for single channel microphone
title_full	Speech enhancement using deep neural network based on mask estimation and harmonic regeneration noise reduction for single channel microphone
title_fullStr	Speech enhancement using deep neural network based on mask estimation and harmonic regeneration noise reduction for single channel microphone
title_full_unstemmed	Speech enhancement using deep neural network based on mask estimation and harmonic regeneration noise reduction for single channel microphone
title_sort	speech enhancement using deep neural network based on mask estimation and harmonic regeneration noise reduction for single channel microphone
granting_institution	Universiti Tun Hussein Onn Malaysia
granting_department	Fakulti Kejuruteraan Elektrik dan Elektronik
publishDate	2022
url	http://eprints.uthm.edu.my/8464/1/24p%20NOREZMI%20MD%20JAMAL.pdf http://eprints.uthm.edu.my/8464/2/NOREZMI%20MD%20JAMAL%20COPYRIGHT%20DECLARATION.pdf http://eprints.uthm.edu.my/8464/3/NOREZMI%20MD%20JAMAL%20WATERMARK.pdf
_version_	1776103352040423424

Speech enhancement using deep neural network based on mask estimation and harmonic regeneration noise reduction for single channel microphone

Similar Items