ISCA Archive Interspeech 2014
ISCA Archive Interspeech 2014

Low-resource open vocabulary keyword search using point process models

Chunxi Liu, Aren Jansen, Guoguo Chen, Keith Kintzley, Jan Trmal, Sanjeev Khudanpur

The point process model (PPM) for keyword search is a whole-word parametric modeling framework based on the timing of phonetic events rather than the evolution of frame-level phonetic likelihoods. Recent progress in PPM training and decoding algorithms has yielded state-of-the-art phonetic search performance in high-resource settings, both in terms of accuracy and computational efficiency. In this paper, we consider PPM application to low-resource settings where the amount of transcribed speech is severely limited and the pronunciation dictionary is incomplete. By using (i) state-of-the-art deep neural network acoustic models to generate phonetic events and (ii) grapheme-to-phoneme conversion to generate pronunciations for out-of-vocabulary (OOV) keywords, we find the PPM system reaches state-of-the-art OOV search performance at a small computational cost. Moreover, due to their complementary methodologies, combining PPM outputs with the LVCSR baseline produces average relative ATWV improvements of 7% and 50% for in-vocabulary and OOV keywords, respectively (16% overall).


doi: 10.21437/Interspeech.2014-533

Cite as: Liu, C., Jansen, A., Chen, G., Kintzley, K., Trmal, J., Khudanpur, S. (2014) Low-resource open vocabulary keyword search using point process models. Proc. Interspeech 2014, 2789-2793, doi: 10.21437/Interspeech.2014-533

@inproceedings{liu14f_interspeech,
  author={Chunxi Liu and Aren Jansen and Guoguo Chen and Keith Kintzley and Jan Trmal and Sanjeev Khudanpur},
  title={{Low-resource open vocabulary keyword search using point process models}},
  year=2014,
  booktitle={Proc. Interspeech 2014},
  pages={2789--2793},
  doi={10.21437/Interspeech.2014-533},
  issn={2308-457X}
}