{"id":156168,"date":"2008-01-01T00:00:00","date_gmt":"2008-01-01T00:00:00","guid":{"rendered":"https:\/\/www.microsoft.com\/en-us\/research\/msr-research-item\/empirical-properties-of-multilingual-phone-to-word-transduction-2\/"},"modified":"2018-10-16T20:16:23","modified_gmt":"2018-10-17T03:16:23","slug":"empirical-properties-of-multilingual-phone-to-word-transduction-2","status":"publish","type":"msr-research-item","link":"https:\/\/www.microsoft.com\/en-us\/research\/publication\/empirical-properties-of-multilingual-phone-to-word-transduction-2\/","title":{"rendered":"Empirical Properties of Multilingual Phone-to-Word Transduction"},"content":{"rendered":"<div class=\"asset-content\">\n<p>This paper explores the error-robustness of phone-to-word transduction across a variety of languages. We implement a noisy channel model in which a phonetic input stream is corrupted by an error model, and then transduced back to words using the inverse error model and linguistic constraints. By controlling the error level, we are able to measure the sensitivity of different languages to degradation in the phonetic input stream. This analysis is carried further to measure the importance of each phone in each language individually. We study Arabic, Chinese, English, German and Spanish, and \ufb01nd that they behave similarly in this paradigm: in each case, a phone error produces about 1.4 word errors, and frequently incorrect phones matter slightly less than others. In the absence of phone errors, transduced word errors are still present, and we use the conditional entropy of words given phones to explain the observed behavior.<\/p>\n<\/div>\n<p><!-- .asset-content --><\/p>\n","protected":false},"excerpt":{"rendered":"<p>This paper explores the error-robustness of phone-to-word transduction across a variety of languages. We implement a noisy channel model in which a phonetic input stream is corrupted by an error model, and then transduced back to words using the inverse error model and linguistic constraints. By controlling the error level, we are able to measure [&hellip;]<\/p>\n","protected":false},"featured_media":0,"template":"","meta":{"msr-url-field":"","msr-podcast-episode":"","msrModifiedDate":"","msrModifiedDateEnabled":false,"ep_exclude_from_search":false,"_classifai_error":"","footnotes":""},"msr-content-type":[3],"msr-research-highlight":[],"research-area":[13545],"msr-publication-type":[193716],"msr-product-type":[],"msr-focus-area":[],"msr-platform":[],"msr-download-source":[],"msr-locale":[268875],"msr-post-option":[],"msr-field-of-study":[],"msr-conference":[],"msr-journal":[],"msr-impact-theme":[],"msr-pillar":[],"class_list":["post-156168","msr-research-item","type-msr-research-item","status-publish","hentry","msr-research-area-human-language-technologies","msr-locale-en_us"],"msr_publishername":"","msr_edition":"In Proceedings of ICASSP","msr_affiliation":"","msr_published_date":"2008-01-01","msr_host":"","msr_duration":"","msr_version":"","msr_speaker":"","msr_other_contributors":"","msr_booktitle":"In Proceedings of ICASSP","msr_pages_string":"","msr_chapter":"","msr_isbn":"","msr_journal":"","msr_volume":"","msr_number":"","msr_editors":"","msr_series":"","msr_issue":"","msr_organization":"","msr_how_published":"","msr_notes":"","msr_highlight_text":"","msr_release_tracker_id":"","msr_original_fields_of_study":"","msr_download_urls":"","msr_external_url":"","msr_secondary_video_url":"","msr_longbiography":"","msr_microsoftintellectualproperty":1,"msr_main_download":"226018","msr_publicationurl":"","msr_doi":"","msr_publication_uploader":[{"type":"file","title":"icassp08a.pdf","viewUrl":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2008\/01\/icassp08a.pdf","id":226018,"label_id":0}],"msr_related_uploader":"","msr_attachments":[{"id":226018,"url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2008\/01\/icassp08a.pdf"}],"msr-author-ordering":[{"type":"user_nicename","value":"gzweig","user_id":31938,"rest_url":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/microsoft-research\/v1\/researchers?person=gzweig"}],"msr_impact_theme":[],"msr_research_lab":[],"msr_event":[],"msr_group":[],"msr_project":[169434,169630],"publication":[],"video":[],"download":[],"msr_publication_type":"inproceedings","related_content":{"projects":[{"ID":169434,"post_title":"Acoustic Modeling","post_name":"acoustic-modeling","post_type":"msr-project","post_date":"2004-01-29 16:42:42","post_modified":"2019-08-14 14:50:04","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/acoustic-modeling\/","post_excerpt":"Acoustic modeling of speech typically refers to the process of\u00a0establishing statistical\u00a0representations for the feature vector sequences\u00a0computed from the speech waveform. Hidden Markov Model (HMM) is one most common type of acoustuc models. Other acosutic models include segmental models, super-segmental models (including hidden dynamic models), neural networks, maximum entropy models, and (hidden) conditional random fields, etc. Acoustic modeling also encompasses \"pronunciation modeling\", which describes how a sequence or multi-sequences of fundamental speech units\u00a0(such as phones or&hellip;","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/169434"}]}},{"ID":169630,"post_title":"Language Modeling for Speech Recognition","post_name":"language-modeling-for-speech-recognition","post_type":"msr-project","post_date":"2004-01-29 16:43:32","post_modified":"2019-08-19 09:41:10","post_status":"publish","permalink":"https:\/\/www.microsoft.com\/en-us\/research\/project\/language-modeling-for-speech-recognition\/","post_excerpt":"Did I just say \"It's fun to recognize speech?\" or \"It's fun to wreck a nice beach?\" It's hard to tell because they sound about the same. Of course, it's a lot more likely that I would say \"recognize speech\" than \"wreck a nice beach.\" Language models help a speech recognizer figure out how likely a word sequence is, independent of the acoustics. This lets the recognizer make the right guess when two different sentences&hellip;","_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-project\/169630"}]}}]},"_links":{"self":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/156168"}],"collection":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item"}],"about":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/types\/msr-research-item"}],"version-history":[{"count":1,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/156168\/revisions"}],"predecessor-version":[{"id":525602,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-item\/156168\/revisions\/525602"}],"wp:attachment":[{"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/media?parent=156168"}],"wp:term":[{"taxonomy":"msr-content-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-content-type?post=156168"},{"taxonomy":"msr-research-highlight","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-research-highlight?post=156168"},{"taxonomy":"msr-research-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/research-area?post=156168"},{"taxonomy":"msr-publication-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-publication-type?post=156168"},{"taxonomy":"msr-product-type","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-product-type?post=156168"},{"taxonomy":"msr-focus-area","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-focus-area?post=156168"},{"taxonomy":"msr-platform","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-platform?post=156168"},{"taxonomy":"msr-download-source","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-download-source?post=156168"},{"taxonomy":"msr-locale","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-locale?post=156168"},{"taxonomy":"msr-post-option","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-post-option?post=156168"},{"taxonomy":"msr-field-of-study","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-field-of-study?post=156168"},{"taxonomy":"msr-conference","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-conference?post=156168"},{"taxonomy":"msr-journal","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-journal?post=156168"},{"taxonomy":"msr-impact-theme","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-impact-theme?post=156168"},{"taxonomy":"msr-pillar","embeddable":true,"href":"https:\/\/www.microsoft.com\/en-us\/research\/wp-json\/wp\/v2\/msr-pillar?post=156168"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}