Speech Enhancement Using Deep Learning Methods: A Review
Speech enhancement, which aims to recover the clean speech of the corrupted signal, plays an important role in the digital speech signal processing. According to the type of degradation and noise in the speech signal, approaches to speech enhancement vary. Thus, the research topic remains challengin...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Indonesian Institute of Sciences
2021-08-01
|
Series: | Jurnal Elektronika dan Telekomunikasi |
Subjects: | |
Online Access: | https://www.jurnalet.com/jet/article/view/392 |
id |
doaj-cbf9deda016e48b38a79883cfdc44667 |
---|---|
record_format |
Article |
spelling |
doaj-cbf9deda016e48b38a79883cfdc446672021-08-31T06:11:22ZengIndonesian Institute of SciencesJurnal Elektronika dan Telekomunikasi1411-82892527-99552021-08-01211192610.14203/jet.v21.19-26218Speech Enhancement Using Deep Learning Methods: A ReviewAsri Rizki Yuliani0M. Faizal Amri1Endang Suryawati2Ade Ramdan3Hilman Ferdinandus Pardede4Research Center for Informatics Indonesian Institute of Sciences (LIPI)Technical Implementation Unit for Instrumental Development Indonesian Institute of Sciences (LIPI)Research Center for Informatics Indonesian Institute of Sciences (LIPI)Research Center for Informatics Indonesian Institute of Sciences (LIPI)Research Center for Informatics Indonesian Institute of Sciences (LIPI)Speech enhancement, which aims to recover the clean speech of the corrupted signal, plays an important role in the digital speech signal processing. According to the type of degradation and noise in the speech signal, approaches to speech enhancement vary. Thus, the research topic remains challenging in practice, specifically when dealing with highly non-stationary noise and reverberation. Recent advance of deep learning technologies has provided great support for the progress in speech enhancement research field. Deep learning has been known to outperform the statistical model used in the conventional speech enhancement. Hence, it deserves a dedicated survey. In this review, we described the advantages and disadvantages of recent deep learning approaches. We also discussed challenges and trends of this field. From the reviewed works, we concluded that the trend of the deep learning architecture has shifted from the standard deep neural network (DNN) to convolutional neural network (CNN), which can efficiently learn temporal information of speech signal, and generative adversarial network (GAN), that utilize two networks training.https://www.jurnalet.com/jet/article/view/392speech enhancementdeep learningneural networksspeech signal processingnon-stationary noise |
collection |
DOAJ |
language |
English |
format |
Article |
sources |
DOAJ |
author |
Asri Rizki Yuliani M. Faizal Amri Endang Suryawati Ade Ramdan Hilman Ferdinandus Pardede |
spellingShingle |
Asri Rizki Yuliani M. Faizal Amri Endang Suryawati Ade Ramdan Hilman Ferdinandus Pardede Speech Enhancement Using Deep Learning Methods: A Review Jurnal Elektronika dan Telekomunikasi speech enhancement deep learning neural networks speech signal processing non-stationary noise |
author_facet |
Asri Rizki Yuliani M. Faizal Amri Endang Suryawati Ade Ramdan Hilman Ferdinandus Pardede |
author_sort |
Asri Rizki Yuliani |
title |
Speech Enhancement Using Deep Learning Methods: A Review |
title_short |
Speech Enhancement Using Deep Learning Methods: A Review |
title_full |
Speech Enhancement Using Deep Learning Methods: A Review |
title_fullStr |
Speech Enhancement Using Deep Learning Methods: A Review |
title_full_unstemmed |
Speech Enhancement Using Deep Learning Methods: A Review |
title_sort |
speech enhancement using deep learning methods: a review |
publisher |
Indonesian Institute of Sciences |
series |
Jurnal Elektronika dan Telekomunikasi |
issn |
1411-8289 2527-9955 |
publishDate |
2021-08-01 |
description |
Speech enhancement, which aims to recover the clean speech of the corrupted signal, plays an important role in the digital speech signal processing. According to the type of degradation and noise in the speech signal, approaches to speech enhancement vary. Thus, the research topic remains challenging in practice, specifically when dealing with highly non-stationary noise and reverberation. Recent advance of deep learning technologies has provided great support for the progress in speech enhancement research field. Deep learning has been known to outperform the statistical model used in the conventional speech enhancement. Hence, it deserves a dedicated survey. In this review, we described the advantages and disadvantages of recent deep learning approaches. We also discussed challenges and trends of this field. From the reviewed works, we concluded that the trend of the deep learning architecture has shifted from the standard deep neural network (DNN) to convolutional neural network (CNN), which can efficiently learn temporal information of speech signal, and generative adversarial network (GAN), that utilize two networks training. |
topic |
speech enhancement deep learning neural networks speech signal processing non-stationary noise |
url |
https://www.jurnalet.com/jet/article/view/392 |
work_keys_str_mv |
AT asririzkiyuliani speechenhancementusingdeeplearningmethodsareview AT mfaizalamri speechenhancementusingdeeplearningmethodsareview AT endangsuryawati speechenhancementusingdeeplearningmethodsareview AT aderamdan speechenhancementusingdeeplearningmethodsareview AT hilmanferdinanduspardede speechenhancementusingdeeplearningmethodsareview |
_version_ |
1721184108959236096 |