Deep Learning Based Automatic Speech Recognition for Turkish

Using Deep Neural Networks (DNN) as an advanced Artificial Neural Networks (ANN) has become widespread with the development of computer technology. Although DNN has been applied for solving Automatic Speech Recognition (ASR) problem in some languages, DNN-based Turkish Speech Recognition has not bee...

Full description

Saved in:
Bibliographic Details
Main Authors: Hamit Erdem, Burak Tombaloğlu
Format: Article
Language:English
Published: Sakarya University 2020-08-01
Series:Sakarya Üniversitesi Fen Bilimleri Enstitüsü Dergisi
Subjects:
Online Access:https://dergipark.org.tr/tr/download/article-file/1210062
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850252236363399168
author Hamit Erdem
Burak Tombaloğlu
author_facet Hamit Erdem
Burak Tombaloğlu
author_sort Hamit Erdem
collection DOAJ
description Using Deep Neural Networks (DNN) as an advanced Artificial Neural Networks (ANN) has become widespread with the development of computer technology. Although DNN has been applied for solving Automatic Speech Recognition (ASR) problem in some languages, DNN-based Turkish Speech Recognition has not been studied extensively. Turkish language is an agglutinative and a phoneme-based language. In this study, a Deep Belief Network (DBN) based Turkish phoneme and speech recognizer is developed. The proposed system recognizes words in the system vocabulary and phoneme components of out of vocabulary (OOV) words. Sub-word (morpheme) based language modelling is implemented into the system. Each phoneme of Turkish language is also modelled as a sub-word in the model. Sub-word (morpheme) based language model is widely used for agglutinative languages to prevent excessive vocabulary size. The performance of the suggested DBN based ASR system is compared with the conventional recognition method, GMM (Gaussian Mixture Method) based Hidden Markov Model (HMM). Regarding to performance metrics, the recognition rate of Turkish language is improved in compare with previous studies.
format Article
id doaj-art-9d062f13260243f79d76ea14c49d8239
institution OA Journals
issn 2147-835X
language English
publishDate 2020-08-01
publisher Sakarya University
record_format Article
series Sakarya Üniversitesi Fen Bilimleri Enstitüsü Dergisi
spelling doaj-art-9d062f13260243f79d76ea14c49d82392025-08-20T01:57:43ZengSakarya UniversitySakarya Üniversitesi Fen Bilimleri Enstitüsü Dergisi2147-835X2020-08-0124472573910.16984/saufenbilder.71188828Deep Learning Based Automatic Speech Recognition for TurkishHamit Erdem0https://orcid.org/0000-0003-1704-1581Burak Tombaloğlu1https://orcid.org/0000-0003-3994-0422BAŞKENT ÜNİVERSİTESİBAŞKENT ÜNİVERSİTESİ, FEN BİLİMLERİ ENSTİTÜSÜUsing Deep Neural Networks (DNN) as an advanced Artificial Neural Networks (ANN) has become widespread with the development of computer technology. Although DNN has been applied for solving Automatic Speech Recognition (ASR) problem in some languages, DNN-based Turkish Speech Recognition has not been studied extensively. Turkish language is an agglutinative and a phoneme-based language. In this study, a Deep Belief Network (DBN) based Turkish phoneme and speech recognizer is developed. The proposed system recognizes words in the system vocabulary and phoneme components of out of vocabulary (OOV) words. Sub-word (morpheme) based language modelling is implemented into the system. Each phoneme of Turkish language is also modelled as a sub-word in the model. Sub-word (morpheme) based language model is widely used for agglutinative languages to prevent excessive vocabulary size. The performance of the suggested DBN based ASR system is compared with the conventional recognition method, GMM (Gaussian Mixture Method) based Hidden Markov Model (HMM). Regarding to performance metrics, the recognition rate of Turkish language is improved in compare with previous studies.https://dergipark.org.tr/tr/download/article-file/1210062deepneuralnetworksbeliefautomaticspeechrecognitionturkishdeep belief networks
spellingShingle Hamit Erdem
Burak Tombaloğlu
Deep Learning Based Automatic Speech Recognition for Turkish
Sakarya Üniversitesi Fen Bilimleri Enstitüsü Dergisi
deep
neural
networks
belief
automatic
speech
recognition
turkish
deep belief networks
title Deep Learning Based Automatic Speech Recognition for Turkish
title_full Deep Learning Based Automatic Speech Recognition for Turkish
title_fullStr Deep Learning Based Automatic Speech Recognition for Turkish
title_full_unstemmed Deep Learning Based Automatic Speech Recognition for Turkish
title_short Deep Learning Based Automatic Speech Recognition for Turkish
title_sort deep learning based automatic speech recognition for turkish
topic deep
neural
networks
belief
automatic
speech
recognition
turkish
deep belief networks
url https://dergipark.org.tr/tr/download/article-file/1210062
work_keys_str_mv AT hamiterdem deeplearningbasedautomaticspeechrecognitionforturkish
AT buraktombaloglu deeplearningbasedautomaticspeechrecognitionforturkish