KIT | KIT-Bibliothek | Impressum

Machine Translation of Spontaneous Speech

Cho, Eunah

Abstract: Recent advances in machine translation have promoted extensive development of many applications. When applied to spontaneous speech input, however, their performance is severely degraded due to the distinctive characteristics of spontaneous speech. In this thesis, machine learning techniques to model the characteristics are developed. It was shown that transforming the spontaneous speech into well-written sentences is crucial in order to improve the machine translation of spontaneous speech.


Zugehörige Institution(en) am KIT Institut für Anthropomatik und Robotik (IAR)
Publikationstyp Hochschulschrift
Jahr 2016
Sprache Englisch
Identifikator DOI(KIT): 10.5445/IR/1000056203
URN: urn:nbn:de:swb:90-562033
KITopen ID: 1000056203
Verlag Karlsruhe
Umfang XVIII, 187 S.
Abschlussart Dissertation
Fakultät Fakultät für Informatik (INFORMATIK)
Institut Institut für Anthropomatik und Robotik (IAR)
Prüfungsdaten 11.02.2016
Referent/Betreuer Prof. A. Waibel
KIT – Die Forschungsuniversität in der Helmholtz-Gemeinschaft KITopen Landing Page