Improving the robustness of automatic speech recognizers using state duration information

N. Nicol, Stephan Euler, M. Falkhausen, Herbert Reininger, Dietrich Wolf, J. Zinke

In this paper we discuss the influence of state duration information on the robustness of an isolated word recognition system operating in noisy environment. Two methods for modeling state durations in Hidden Markov Models are compared. First, a method for modeling the distribution of state durations with Poisson statistics and second, a method of internal state duration modeling. Recognition results obtained with a vocabulary of 23 German words disturbed with two different kinds of noise are presented.

Cite as: Nicol, N., Euler, S., Falkhausen, M., Reininger, H., Wolf, D., Zinke, J. (1992) Improving the robustness of automatic speech recognizers using state duration information. Proc. ETRW on Speech Processing in Adverse Conditions, 183-186

@inproceedings{nicol92_spac,
  author={N. Nicol and Stephan Euler and M. Falkhausen and Herbert Reininger and Dietrich Wolf and J. Zinke},
  title={{Improving the robustness of automatic speech recognizers using state duration information}},
  year=1992,
  booktitle={Proc. ETRW on Speech Processing in Adverse Conditions},
  pages={183--186}
}