ISCA Archive Interspeech 2008
ISCA Archive Interspeech 2008

Context-dependent phone models and models adaptation for phonotactic language recognition

Mohamed Faouzi BenZeghiba, Jean-Luc Gauvain, Lori Lamel

The performance of a PPRLM language recognition system depends on the quality and the consistency of phone decoders. To improve the performance of the decoders, this paper investigates the use of context-dependent instead of context-independent phone models, and the use of CMLLR for model adaptation. This paper also discusses several improvements to the LIMSI 2007 NIST LRE system, including the use of a 4-gram language model, score calibration and fusion using the FoCal Multi-class toolkit (with large development data) and better decoding parameters such as phone insertion penalty. The improved system is evaluated on the NIST LRE-2005 and the LRE-2007 evaluation data sets. Despite its simplicity, the system achieves for the 30s condition a Cavg of 2.4% and 1.6% on these data sets, respectively.


doi: 10.21437/Interspeech.2008-142

Cite as: BenZeghiba, M.F., Gauvain, J.-L., Lamel, L. (2008) Context-dependent phone models and models adaptation for phonotactic language recognition. Proc. Interspeech 2008, 313-316, doi: 10.21437/Interspeech.2008-142

@inproceedings{benzeghiba08_interspeech,
  author={Mohamed Faouzi BenZeghiba and Jean-Luc Gauvain and Lori Lamel},
  title={{Context-dependent phone models and models adaptation for phonotactic language recognition}},
  year=2008,
  booktitle={Proc. Interspeech 2008},
  pages={313--316},
  doi={10.21437/Interspeech.2008-142}
}