ISCA Archive ICSLP 1994
ISCA Archive ICSLP 1994

Integrating TDNN-based diphone recognition with table-driven morphology parsing for understanding of spoken Korean

Kyunghee Kim, Geunbae Lee, Jong-Hyeok Lee, Hong Jeong

In this paper, we propose a spoken Korean morphological analysis model extensible to large vocabulary continuous speech recognition. This model consists of a diphone recognizer, a diphone2phoneme filter and a CYK-morphological analyzer. Two-level hierarchical TDNNs (time-delayed neural networks) recognize Korean diphones which are transformed into a phoneme lattice (a set of phoneme candidates hypothesized by a speech recognition module) by a diphone2phoneme filter. The morphological analyzer parses the phoneme lattice with the phonological changes handling and produces the morphology-segmented Korean words (called Eojeols). Using the TDNN diphone speech recognizer, we obtained 95.2% of 17 Korean vowel recognition and 93.7% of 72 diphone recognition. The speaker-dependent and continuous Eojeol recognition experiments using the current model show that the morphological analysis for spoken Korean can be achieved for medium sized vocabularies with 90.6% of success rate.


Cite as: Kim, K., Lee, G., Lee, J.-H., Jeong, H. (1994) Integrating TDNN-based diphone recognition with table-driven morphology parsing for understanding of spoken Korean. Proc. 3rd International Conference on Spoken Language Processing (ICSLP 1994), 5-8

@inproceedings{kim94_icslp,
  author={Kyunghee Kim and Geunbae Lee and Jong-Hyeok Lee and Hong Jeong},
  title={{Integrating TDNN-based diphone recognition with table-driven morphology parsing for understanding of spoken Korean}},
  year=1994,
  booktitle={Proc. 3rd International Conference on Spoken Language Processing (ICSLP 1994)},
  pages={5--8}
}