ISCA Archive Interspeech 2009
ISCA Archive Interspeech 2009

Tied-state multi-path HMnet model using three-domain successive state splitting

Soo-Young Suk, Hiroaki Kojima

In this paper, we address the improvement of an acoustic model using the multi-path Hidden Markov network (HMnet) model for automatically creating non-uniform tied-state, context-dependent hidden markov model topologies. Recent research has achieved multi-path model topologies in order to improve the recognition performance in gender-independent, spontaneous-speaking applications. However, the multi-path acoustic model size may increase and require more training samples depending on the increased number of paths. To solve this problem, we used a tied-state multipath topology by which we can create a three-domain successive state splitting method to which environmental splitting is added. This method can obtain a suitable model topology with small mixture components. Experiments demonstrated that the proposed multi-path HMnet model performs better than single-path models for the same number of states.


doi: 10.21437/Interspeech.2009-436

Cite as: Suk, S.-Y., Kojima, H. (2009) Tied-state multi-path HMnet model using three-domain successive state splitting. Proc. Interspeech 2009, 1419-1422, doi: 10.21437/Interspeech.2009-436

@inproceedings{suk09_interspeech,
  author={Soo-Young Suk and Hiroaki Kojima},
  title={{Tied-state multi-path HMnet model using three-domain successive state splitting}},
  year=2009,
  booktitle={Proc. Interspeech 2009},
  pages={1419--1422},
  doi={10.21437/Interspeech.2009-436}
}