{"id":{"repo_id":"uiuc","oai_identifier":"oai:www.ideals.illinois.edu:2142/113890"},"canonical_url":"https://search.dev.ndltd.org/etd/uiuc/oai:www.ideals.illinois.edu:2142/113890","repository":{"repo_id":"uiuc","name":"University of Illinois - Urbana-Champaign","base_url":"https://www.ideals.illinois.edu/oai-pmh"},"display":{"title":"Improving multilingual speech recognition systems","abstract":"Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-04-06 without embargo terms","abstract_html":"Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-04-06 without embargo terms","abstract_has_math":false,"creators":["Gao, Heting"],"institution":"University of Illinois at Urbana-Champaign","degree_name":"M.S.","degree_level":"Thesis","degree_discipline":"Electrical & Computer Engr","degree_department":null,"school":null,"contributors":["Hasegawa-Johnson, Mark"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2022,"date_issued":"2022-04-29T21:34:45Z","date_published":"2022-04-29T21:34:45Z","updated_at":"2026-07-22T22:24:53Z","subjects":["Engineering"],"languages":["en","eng"],"rights":["Copyright 2021 Heting Gao"],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"http://hdl.handle.net/2142/113890","outbound_label":"Handle","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Hasegawa-Johnson, Mark"]},{"key":"dc:creator","label":"Author","values":["Gao, Heting"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date","label":"Dc Date","values":["2022-04-29T21:34:45Z","2021-12","2021-12-03"]},{"key":"dc:type","label":"Dc Type","values":["text","Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Electrical & Computer Engr"]},{"key":"thesis:degree_level","label":"Degree Level","values":["Thesis"]},{"key":"thesis:degree_name","label":"Degree Name","values":["M.S."]},{"key":"thesis:institution_name","label":"Thesis Institution Name","values":["University of Illinois at Urbana-Champaign"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Engineering"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language","label":"Dc Language","values":["en","eng"]},{"key":"dc:rights","label":"Dc Rights","values":["Copyright 2021 Heting Gao"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["http://hdl.handle.net/2142/113890"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description","label":"Description","values":["Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-04-06 without embargo terms","The student, Heting Gao, accepted the attached license on 2021-12-01 at 10:37.","The student, Heting Gao, submitted this Thesis for approval on 2021-12-01 at 10:52.","This Thesis was approved for publication on 2021-12-03 at 14:12.","DSpace SAF Submission Ingestion Package generated from Vireo submission #17329 on 2022-04-06 at 17:10:32","Made available in DSpace on 2022-04-29T21:34:45Z (GMT). No. of bitstreams: 2 GAO-THESIS-2021.pdf: 894831 bytes, checksum: 1ee58739f01711f512929df6db9d6be0 (MD5) LICENSE.txt: 4207 bytes, checksum: 6e1f7a541388abeb60a4e4428a13409a (MD5) Previous issue date: 2021-12-03","End-to-end trainable deep neural networks have become the state-of-the-art architecture for automatic speech recognition (ASR), provided that the network is trained with a sufficiently large dataset. However, many existing languages are too sparsely resourced for deep learning networks to achieve as high accuracy as their resource-abundant counterparts. Multilingual recognition systems mitigate data sparsity issues by training models on data from multiple language resources to learn a speech-to-text or speech-to-phone model universal to all languages. The resulting multilingual ASR models usually have better recognition accuracy than the models trained on the individual dataset. In this work, we propose that two limitations exist for multilingual systems, and resolving the two limitations could result in improved recognition accuracy: (1) existing corpora are of the considerably varied form (spontaneous or read speech), corpus size, noise level, and phoneme distribution and the ASR models trained on the joint multilingual dataset have large performance disparities over different languages. We present an optimizable loss function, equal accuracy ratio (EAR), that measures the sequence-level performance disparity between different user groups and we show that explicitly optimizing this objective reduces the performance gap and improves the multilingual recognition accuracy. (2) While having good accuracy on the seen training language, the multilingual systems do not generalize well to unseen testing languages, which we refer to as cross-lingual recognition accuracy. We introduce language embedding using external linguistic typologies and show that such embedding can significantly increase both multilingual and cross-lingual accuracy. We illustrate the effectiveness of the proposed methods with experiments on multilingual and multi-user and multi-dialect corpora."]},{"key":"dc:format","label":"Dc Format","values":["application/pdf"]},{"key":"dc:title","label":"Title","values":["Improving multilingual speech recognition systems"]}]}],"canonical_facts":{"dc:contributor":["Hasegawa-Johnson, Mark"],"dc:creator":["Gao, Heting"],"dc:date":["2022-04-29T21:34:45Z","2021-12","2021-12-03"],"dc:description":["Submission original under an indefinite embargo labeled 'Open Access'. The submission was exported from vireo on 2022-04-06 without embargo terms","The student, Heting Gao, accepted the attached license on 2021-12-01 at 10:37.","The student, Heting Gao, submitted this Thesis for approval on 2021-12-01 at 10:52.","This Thesis was approved for publication on 2021-12-03 at 14:12.","DSpace SAF Submission Ingestion Package generated from Vireo submission #17329 on 2022-04-06 at 17:10:32","Made available in DSpace on 2022-04-29T21:34:45Z (GMT). No. of bitstreams: 2 GAO-THESIS-2021.pdf: 894831 bytes, checksum: 1ee58739f01711f512929df6db9d6be0 (MD5) LICENSE.txt: 4207 bytes, checksum: 6e1f7a541388abeb60a4e4428a13409a (MD5) Previous issue date: 2021-12-03","End-to-end trainable deep neural networks have become the state-of-the-art architecture for automatic speech recognition (ASR), provided that the network is trained with a sufficiently large dataset. However, many existing languages are too sparsely resourced for deep learning networks to achieve as high accuracy as their resource-abundant counterparts. Multilingual recognition systems mitigate data sparsity issues by training models on data from multiple language resources to learn a speech-to-text or speech-to-phone model universal to all languages. The resulting multilingual ASR models usually have better recognition accuracy than the models trained on the individual dataset. In this work, we propose that two limitations exist for multilingual systems, and resolving the two limitations could result in improved recognition accuracy: (1) existing corpora are of the considerably varied form (spontaneous or read speech), corpus size, noise level, and phoneme distribution and the ASR models trained on the joint multilingual dataset have large performance disparities over different languages. We present an optimizable loss function, equal accuracy ratio (EAR), that measures the sequence-level performance disparity between different user groups and we show that explicitly optimizing this objective reduces the performance gap and improves the multilingual recognition accuracy. (2) While having good accuracy on the seen training language, the multilingual systems do not generalize well to unseen testing languages, which we refer to as cross-lingual recognition accuracy. We introduce language embedding using external linguistic typologies and show that such embedding can significantly increase both multilingual and cross-lingual accuracy. We illustrate the effectiveness of the proposed methods with experiments on multilingual and multi-user and multi-dialect corpora."],"dc:format":["application/pdf"],"dc:identifier":["http://hdl.handle.net/2142/113890"],"dc:language":["en","eng"],"dc:rights":["Copyright 2021 Heting Gao"],"dc:subject":["Engineering"],"dc:title":["Improving multilingual speech recognition systems"],"dc:type":["text","Thesis"],"thesis:degree_discipline":["Electrical & Computer Engr"],"thesis:degree_level":["Thesis"],"thesis:degree_name":["M.S."],"thesis:institution_name":["University of Illinois at Urbana-Champaign"]},"updated_at":"2026-07-22T22:24:53Z"}