The 2014 summer workshop is dedicated to the memory of Professor Frederick Jelinek (1932-2010) and is the capstone event of his National Science Foundation award from the International Science and Engineering Section of the Office of International and Integrative Activities under the Partnerships for International Research and Education (PIRE) program.
Fred’s PIRE project links senior and junior researchers from Johns Hopkins University and Brown University with counterparts from Charles University in Prague and from Brno University Technology, both in the Czech Republic, Saarland University in Germany, and Macquarie University in Australia. This international team is investigating formal representations of linguistic meaning for use in automatic speech recognition/reconstruction (ASR) and machine translation (MT) systems. Their goal is to augment current ASR and MT systems by applying a variety of formal models for deep syntactic/semantic representation so that outputs of their refined systems become more accurate, coherent and grammatical.
The PIRE project’s complementary education component involves introducing participating graduate students to non-US linguistic formalisms by giving them the opportunity to work and interact with their international counterparts, and training the students to apply those formalisms to problems in speech and natural language processing. Additional travel by European/Australian researchers to American universities further facilitates this exchange.
The Summer Workshop component of the PIRE project will bring together three recurring themes that run through this project: improved recognition of conversational speech, probabilistic representations of linguistic meaning, and abstract meaning representations for machine translation.
Through working intensively alongside each other over the summer at the historic Prague address that now houses Charles University’s Institute for Formal and Applied Linguistics and was Fred Jelinek’s second academic home, the US and International participants will gain a rich but unifying understanding of the diverse facets of “meaning” as it is manifested in language, and endeavor to apply it in machine processing of spoken and written language.