{"type":"ScholarlyArticle","title":"Repairing ASR output by Artificial Development and Ontology based Learning","authors":[{"name":"C. Anantaram","slug":"c-anantaram","given":"C.","family":"Anantaram","affiliation":"TCS Innovation Labs , Tata Consultancy Services Limited , Gurgaon , India"},{"name":"Amit Sangroya","slug":"amit-sangroya","given":"Amit","family":"Sangroya","affiliation":"TCS Innovation Labs , Tata Consultancy Services Limited , Gurgaon , India"},{"name":"Mrinal Rawat","slug":"mrinal-rawat","given":"Mrinal","family":"Rawat","affiliation":"TCS Innovation Labs , Tata Consultancy Services Limited , Gurgaon , India"},{"name":"Aishwarya Chhabra","slug":"aishwarya-chhabra","given":"Aishwarya","family":"Chhabra","affiliation":"TCS Innovation Labs , Tata Consultancy Services Limited , Gurgaon , India"}],"abstract":"General purpose automatic speech recognition\r\n(gpASR) systems such as Google, Watson, etc.\r\nsometimes output inaccurate sentences when used\r\nin a domain specific scenario as it may not have had\r\nenough training samples for that particular domain\r\nand context. Further, the accent of the speaker and\r\nthe environmental conditions in which the speaker\r\nspeaks a sentence may influence the speech engine\r\nto recognize certain words inaccurately. Many approaches\r\nto improve the accuracy of ASR output\r\nexist. However, in the context of a domain and\r\nthe environment in which a speaker speaks the sentences,\r\ngpASR output needs a lot of improvement\r\nin order to provide effective speech interfaces to\r\ndomain-specific systems. In this paper, we demonstrate\r\na method that combines bio-inspired artifi-\r\ncial development (ArtDev) with machine learning\r\n(ML) approaches to repair the output of a gpASR.\r\nOur method factors in the environment to tailor the\r\nrepair process.","doi":"10.24963/ijcai.2018/842","url":"https://ijcai.org/proceedings/2018/842","pdf":"https://ijcai.org/proceedings/2018/0842.pdf","markdown":"https://ijcai.org/proceedings/2018/842.md","bibtex":"https://ijcai.org/proceedings/2018/bibtex/842","volume":{"slug":"2018","title":"Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2018","url":"https://ijcai.org/proceedings/2018/","year":2018,"location":"Stockholm","dates":"13-19 July 2018","editors":"Jérôme Lang","isbn":"978-0-9992411-2-7","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2018"},"section":"Demos","track":"Demos","number":842,"pages":{"first":5799,"last":5801},"keywords":["Machine Learning: Machine Learning","Knowledge Representation, Reasoning, and Logic: Reasoning about Knowlege and Belief","Natural Language Processing: Natural Language Processing","Natural Language Processing: Speech"],"videos":[],"references":[{"position":1,"raw":"[Anantaram et al., 2016] C. Anantaram, Sunil Kumar Kop- parapu, Chiragkumar Patel, and Aditya Mittal. Repairing general-purpose ASR output to improve accuracy of spo- ken sentences in specific domains using artificial develop- ment approach. In Proceedings of the Twenty-Fifth Inter- national Joint Conference on Artificial Intelligence, IJCAI 2016, New York, NY, USA, 9-15 July 2016, pages 4234- 4235, 2016.","title":"Repairing general-purpose ASR output to improve accuracy of spoken sentences in specific domains using artificial development approach","authors":"C Anantaram, Sunil Kumar Kopparapu, Chiragkumar Patel, Aditya Mittal","year":"2016","venue":"Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence, IJCAI 2016","doi":null,"url":null,"internal_url":"/proceedings/2016/637"},{"position":2,"raw":"[Fusayasu et al., 2015a] Yohei Fusayasu, Katsuyuki Tanaka, Tetsuya Takiguchi, and Yasuo Ariki. Word-error correc- tion of continuous speech recognition based on normalized relevance distance. In Yang and Wooldridge [2015], pages 1257-1262.","title":"Word-error correction of continuous speech recognition based on normalized relevance distance","authors":"Fusayasu","year":"2015","venue":null,"doi":null,"url":null,"internal_url":"/proceedings/2015/181"},{"position":3,"raw":"[Fusayasu et al., 2015b] Yohei Fusayasu, Katsuyuki Tanaka, Tetsuya Takiguchi, and Yasuo Ariki. Word-error correc- tion of continuous speech recognition based on normalized relevance distance. In Yang and Wooldridge [2015], pages 1257-1262.","title":"Word-error correction of continuous speech recognition based on normalized relevance distance","authors":"Fusayasu","year":"2015","venue":null,"doi":null,"url":null,"internal_url":"/proceedings/2015/181"},{"position":4,"raw":"[Gurunath Shivakumar et al., 2018] P. Gurunath Shivaku- mar, H. Li, K. Knight, and P. Georgiou. Learning from Past Mistakes: Improving Automatic Speech Recognition Output via Noisy-Clean Phrase Context Modeling. ArXiv e-prints, February 2018.","title":"Learning from Past Mistakes: Improving Automatic Speech Recognition Output via Noisy-Clean Phrase Context Modeling","authors":"Gurunath Shivakumar","year":"2018","venue":null,"doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Harding and Banzhaf, 2008] Simon Harding and Wolfgang Banzhaf. Artificial Development, pages 201-219. Springer Berlin Heidelberg, Berlin, Heidelberg, 2008.","title":"Artificial Development","authors":"Banzhaf Harding","year":"2008","venue":null,"doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Kim et al., 2016] Byeongchang Kim, Junhwi Choi, and Gary Geunbae Lee. Asr error management using rnn based syllable prediction for spoken dialog applications.","title":"Asr error management using rnn based syllable prediction for spoken dialog applications","authors":"Kim","year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":7,"raw":"In James J. (Jong Hyuk) Park, Gangman Yi, Young-Sik Jeong, and Hong Shen, editors, Advances in Parallel and Distributed Computing and Ubiquitous Services, pages 99-106, Singapore, 2016. Springer Singapore.","title":"Advances in Parallel and Distributed Computing and Ubiquitous Services","authors":null,"year":"2016","venue":null,"doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Morbini et al., 2013] Fabrizio Morbini, Kartik Audhkhasi, Kenji Sagae, Ron Artstein, Dogan Can, Panayiotis Geor- giou, Shri Narayanan, Anton Leuski, and David Traum. Which asr should i choose for my dialogue system? In Proceedings of the SIGDIAL 2013 Conference, pages 394- 403. Association for Computational Linguistics, 2013.","title":"Which asr should i choose for my dialogue system?","authors":"Fabrizio Morbini, Kartik Audhkhasi, Kenji Sagae, Ron Artstein, Dogan Can, Panayiotis Georgiou, Shri Narayanan, Anton Leuski, David Traum","year":"2013","venue":"Proceedings of the SIGDIAL 2013 Conference","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Peng et al., 2013] F. Peng, S. Roy, B. Shahshahani, and F. Beaufays. Search results based n-best hypothesis rescor- ing with maximum entropy classification. In 2013 IEEE Workshop on Automatic Speech Recognition and Under- standing, pages 422-427, Dec 2013.","title":"Search results based n-best hypothesis rescoring with maximum entropy classification","authors":"F Peng, S Roy, B Shahshahani, F Beaufays","year":"2013","venue":"IEEE Workshop on Automatic Speech Recognition and Understanding","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Suhm et al., 2001] Bernhard Suhm, Brad Myers, and Alex Waibel. Multimodal error correction for speech user inter- faces. ACM Trans. Comput.-Hum. Interact., 8(1):60-98, March 2001.","title":"Multimodal error correction for speech user interfaces","authors":"Bernhard Suhm, Brad Myers, Alex Waibel","year":"2001","venue":"ACM Trans. Comput.-Hum. Interact","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Tufte, 2009] Gunnar Tufte. From evo to evodevo: Mapping and adaptation in artificial development. In Wellington Pinheiro dos Santos, editor, Evolutionary Computation, chapter 12. InTech, Rijeka, 2009.","title":"From evo to evodevo: Mapping and adaptation in artificial development","authors":"Gunnar Tufte, Tufte","year":"2009","venue":"Evolutionary Computation","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Tür et al., 2013] Gökhan Tür, Anoop Deoras, and Dilek Hakkani-Tür. Semantic parsing using word confusion net- works with conditional random fields. In Frédéric Bimbot, Christophe Cerisara, Cécile Fougeron, Guillaume Gravier, Lori Lamel, Franc ¸ois Pellegrino, and Pascal Perrier, edi- tors, INTERSPEECH 2013, 14th Annual Conference of the International Speech Communication Association, Lyon, France, August 25-29, 2013, pages 2579-2583. ISCA, 2013.","title":"Semantic parsing using word confusion networks with conditional random fields","authors":"Gökhan Tür, Anoop Deoras, Dilek Hakkani-Tür","year":"2013","venue":"INTERSPEECH 2013, 14th Annual Conference of the International Speech Communication Association","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Yang and Wooldridge, 2015] Qiang Yang and Michael Wooldridge, editors. Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence, IJCAI 2015, Buenos Aires, Argentina, July 25-31, 2015. AAAI Press, 2015.","title":"Proceedings of the Twenty-Fourth International Joint Conference on Artificial Intelligence, IJCAI 2015","authors":null,"year":"2015","venue":null,"doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}