000002792 001__ 2792 000002792 005__ 20251203131633.0 000002792 0247_ $$2DOI$$a10.6083/M48C9V0V 000002792 037__ $$aETD 000002792 245__ $$aDiscriminative joint modeling of acoustic and lexical variations for spoken language processing 000002792 260__ $$bOregon Health and Science University 000002792 269__ $$a2014 000002792 336__ $$aDissertation 000002792 502__ $$bPh.D. 000002792 502__ $$gComputer Science & Electrical Engineering (sunsetting) 000002792 520__ $$aSpeech recognition systems consist of three components, namely, the acoustic model, the pronunciation model and the language model. The acoustic and language models are typically learned separately and furthermore optimized for different cost functions. This framework has been a result of historical and practical considerations such as the availabil- ity of limited amounts of training data and the computational cost. These considerations are currently being overcome. Arguably, learning both models jointly to directly minimize the word error rate will result in a better recognizer. One of the contributions of this thesis is a detailed investigation of a discriminative framework to jointly learn the parameters of the acoustic, language and duration models (commonly captured with the parameters from the acoustic models). 000002792 540__ $$fCC BY 000002792 542__ $$fIn copyright - single owner 000002792 650__ $$aSpeech Recognition Software$$036177 000002792 650__ $$aTransducers$$027257 000002792 650__ $$aAcoustics$$014161 000002792 650__ $$aLanguage$$021309 000002792 650__ $$aSpeech Perception$$026233 000002792 650__ $$aSpeech$$026226 000002792 691__ $$aSchool of Medicine$$041369 000002792 692__ $$aCenter for Spoken Language Understanding$$041388 000002792 7001_ $$aLehr, Maider$$uOregon Health and Science University$$041354 000002792 7201_ $$aShafran, Izhak$$uGoogle, Inc.$$7Personal$$eAdvisor 000002792 8564_ $$969022b53-2645-4bde-9e39-85cb452f3caf$$s2786424$$uhttps://digitalcollections.ohsu.edu/record/2792/files/3560_etd.pdf$$ePublic$$236ff767339e3c0b40776802cfccce352$$31 000002792 905__ $$a/rest/prod/dn/39/x1/69/dn39x1692 000002792 909CO $$ooai:digitalcollections.ohsu.edu:2792$$pstudent-work 000002792 980__ $$aTheses and Dissertations