{"created":"2020-09-01T14:30:08.791636+00:00","id":4316,"links":{},"metadata":{"_buckets":{"deposit":"2f050524-0f3d-4eab-bbde-45d41b02b92a"},"_deposit":{"id":"4316","owners":[],"pid":{"revision_id":0,"type":"recid","value":"4316"},"status":"published"},"_oai":{"id":"oai:meral.edu.mm:recid/4316","sets":["1582963302567:1597824322519"]},"communities":["ucsy"],"item_1583103067471":{"attribute_name":"Title","attribute_value_mlt":[{"subitem_1551255647225":"Myanmar Language Continuous Speech Recognition Using Convolutional Neural Network (CNN)","subitem_1551255648112":"en_US"}]},"item_1583103085720":{"attribute_name":"Description","attribute_value_mlt":[{"interim":"Researchers of many nations have developed automatic speech recognition(ASR) to show their national improvement in information and communicationtechnology for their languages.The dissertation aims to develop good quality Myanmar language automaticspeech recognition on read speech. Myanmar language is being considered as a lowresourced language. Thus, there is no speech corpus which is freely and commerciallyavailable for ASR research. Therefore, a speech corpus named “University ofComputer Studies Yangon - Speech Corpus (UCSY-SC1)” which is essential forMyanmar ASR research is constructed. The speech corpus is developed by using twotypes of domains: web news and daily conversations. The news is collected from theInternet and the conversational data is recorded by ourselves. This corpus is applied tobuild the Myanmar ASR.Myanmar language is one of the tonal languages and different types of tonesconvey the difference in meanings. Therefore, like the other tonal languages such asMandarin, Vietnamese and Thai, tone information is significantly played to improvethe Myanmar ASR performance. Moreover, syllable is the basic unit of Myanmarlanguage. Thus, in this work, the effect of tones is explored on both syllable andword-based ASR models. The comparison of syllable-based ASR model and wordbased ASR model is also done.In this work, Myanmar ASR is built by applying state-of-the-art acousticmodel, Convolutional Neural Network (CNN). In low-resourced condition, CNN isbetter than Deep Neural Network (DNN) because the fully connected nature of theDNN can cause overfitting. And it degrades the ASR performance for low-resourcedlanguages where there is a limited amount of training data. CNN can alleviate theseproblems and it is very useful for a low-resourced language such as Myanmar.Furthermore, CNN can model well tone patterns because it can reduce spectralvariations and model spectral correlations existing in the signal. In this task, it showedthat CNN outperformed DNN and Gaussian Mixture Model (GMM)-Hidden MarkovModel (HMM). The best accuracy is achieved with CNN-based model in MyanmarASR."}]},"item_1583103108160":{"attribute_name":"Keywords","attribute_value":[]},"item_1583103120197":{"attribute_name":"Files","attribute_type":"file","attribute_value_mlt":[{"accessrole":"open_access","date":[{"dateType":"Available","dateValue":"2019-09-23"}],"displaytype":"preview","filename":"AyeNyeinMonThesisBook.pdf","filesize":[{"value":"4614 Kb"}],"format":"application/pdf","licensetype":"license_note","mimetype":"application/pdf","url":{"url":"https://meral.edu.mm/record/4316/files/AyeNyeinMonThesisBook.pdf"},"version_id":"5227c730-6d67-4f80-bbcf-26984c2a25c1"}]},"item_1583103131163":{"attribute_name":"Journal articles","attribute_value_mlt":[{"subitem_issue":"","subitem_journal_title":"","subitem_pages":"","subitem_volume":""}]},"item_1583103147082":{"attribute_name":"Conference papers","attribute_value_mlt":[{"subitem_acronym":"","subitem_c_date":"","subitem_conference_title":"","subitem_part":"","subitem_place":"","subitem_session":"","subitem_website":""}]},"item_1583103211336":{"attribute_name":"Books/reports/chapters","attribute_value_mlt":[{"subitem_book_title":"","subitem_isbn":"","subitem_pages":"","subitem_place":"","subitem_publisher":""}]},"item_1583103233624":{"attribute_name":"Thesis/dissertations","attribute_value_mlt":[{"subitem_awarding_university":"University of Computer Studies, Yangon","subitem_supervisor(s)":[{"subitem_supervisor":""}]}]},"item_1583105942107":{"attribute_name":"Authors","attribute_value_mlt":[{"subitem_authors":[{"subitem_authors_fullname":"Mon, Aye Nyein"}]}]},"item_1583108359239":{"attribute_name":"Upload type","attribute_value_mlt":[{"interim":"Publication"}]},"item_1583108428133":{"attribute_name":"Publication type","attribute_value_mlt":[{"interim":"Thesis"}]},"item_1583159729339":{"attribute_name":"Publication date","attribute_value":"2019-01"},"item_1583159847033":{"attribute_name":"Identifier","attribute_value":"http://onlineresource.ucsy.edu.mm/handle/123456789/2255"},"item_title":"Myanmar Language Continuous Speech Recognition Using Convolutional Neural Network (CNN)","item_type_id":"21","owner":"1","path":["1597824322519"],"publish_date":"2019-09-23","publish_status":"0","recid":"4316","relation_version_is_last":true,"title":["Myanmar Language Continuous Speech Recognition Using Convolutional Neural Network (CNN)"],"weko_creator_id":"1","weko_shared_id":-1},"updated":"2021-12-13T00:54:23.814618+00:00"}