This paper proposes a classification of postpositions for spoken Japanese speech recognition based on empirical data. The speaker's attitude toward his utterance is best expressed in the sentence final phrase in Japanese. Therefore, we first retrieve our empirical database to investigate the syntactic varieties in the sentence final phrase, especially focusing on the conjunctive postpositions. Second, we investigate how they behave in speech recognition by applying loosely constrained syntactic rules. Further, we refine the rules, based on the study. Finally, we establish the validity of their improvement by both speech recognition rate and acceptance rate. Keywords: Spoken Japanese, Postposition, Empirical data, Speech recognition
Cite as: Hosaka, J., Takezawa, T., Ehara, T. (1991) Utilizing empirical data for postposition classification toward spoken Japanese speech recognition. Proc. 2nd European Conference on Speech Communication and Technology (Eurospeech 1991), 573-576, doi: 10.21437/Eurospeech.1991-147
@inproceedings{hosaka91_eurospeech,
author={Junko Hosaka and Toshiyuki Takezawa and Terumasa Ehara},
title={{Utilizing empirical data for postposition classification toward spoken Japanese speech recognition}},
year=1991,
booktitle={Proc. 2nd European Conference on Speech Communication and Technology (Eurospeech 1991)},
pages={573--576},
doi={10.21437/Eurospeech.1991-147}
}