000009298 001__ 9298
000009298 005__ 20240124114320.0
000009298 0247_ $$2DOI$$a10.6083/xk81jm08r
000009298 037__ $$aETD
000009298 245__ $$aImproving speech intelligibility through spectral style conversion
000009298 260__ $$bOregon Health and Science University
000009298 269__ $$a2021
000009298 336__ $$aThesis
000009298 502__ $$bPh.D.
000009298 520__ $$aOral communication is the most important way for delivering information in our daily life. Unfortu-nately, the quality of such communication can be degraded by 1) speech disorders (e.g. dysarthria) and 2) surrounding environments (e.g. noise or reverberation). Style conversion is a technology that modifies the source speaking style of a speaker to sound like a more intelligible target speak-ing style of either the same or different speaker. In the dissertation, I consider new machine learning based-approaches for style conversion. Inspired by the intelligibility gain of clear (CLR) speaking style over habitual (HAB) speaking style, I propose several HAB-to-CLR spectral mappings approaches for intelligibility improvement.
000009298 650__ $$aDysarthria$$018128
000009298 650__ $$aSpeech Intelligibility$$026231
000009298 650__ $$aMachine Learning$$011449
000009298 650__ $$aParkinson Disease$$023617
000009298 6531_ $$acomputer neural networks
000009298 6531_ $$aalaryngeal speech
000009298 691__ $$aSchool of Medicine$$041369
000009298 7001_ $$aDinh, Tuan A.
000009298 8564_ $$96706c31e-4275-4503-b7f5-8f73be7b8004$$s5342506$$uhttps://digitalcollections.ohsu.edu/record/9298/files/Dinh.Tuan.2021.pdf
000009298 905__ $$a/rest/prod/xk/81/jm/08/xk81jm08r
000009298 909CO $$ooai:digitalcollections.ohsu.edu:9298$$pstudent-work
000009298 980__ $$aTheses and Dissertations