Reducing Confusions and Ambiguities in Speech Translation - 2005

Explore cutting-edge research on improving speech translation in this 1-hour 12-minute lecture by Pascale Fung from the Hong Kong University of Science and Technology. Delve into techniques for reducing acoustic-phonetic confusions in accented Mandarin speech, extracting named entities from spoken language, and applying semantics-based translation disambiguation. Learn how these approaches can significantly enhance speech translation accuracy, even in cases of imperfect speech recognition. Discover the importance of addressing accent-related challenges, understanding the "What, Who, When, Where, Why" of spoken content, and leveraging frame semantics for more precise translations. Gain insights into the latest advancements in speech translation technology and their potential impact on cross-lingual communication.

Reducing Confusions and Ambiguities in Speech Translation – Pascale Fung (HKUST) - 2005