ISCA Archive Interspeech 2008
ISCA Archive Interspeech 2008

Comparing word, character, and phoneme n-grams for subjective utterance recognition

Theresa Wilson, Stephan Raaijmakers

In this paper, we compare the performance of classifiers trained using word n-grams, character n-grams, and phoneme n-grams for recognizing subjective utterances in multiparty conversation. We show that there is value in using very shallow linguistic representations, such as character n-grams, for recognizing subjective utterances, in particular, gains in the recall of subjective utterances.


doi: 10.21437/Interspeech.2008-270

Cite as: Wilson, T., Raaijmakers, S. (2008) Comparing word, character, and phoneme n-grams for subjective utterance recognition. Proc. Interspeech 2008, 1614-1617, doi: 10.21437/Interspeech.2008-270

@inproceedings{wilson08b_interspeech,
  author={Theresa Wilson and Stephan Raaijmakers},
  title={{Comparing word, character, and phoneme n-grams for subjective utterance recognition}},
  year=2008,
  booktitle={Proc. Interspeech 2008},
  pages={1614--1617},
  doi={10.21437/Interspeech.2008-270}
}