ISCA Archive Eurospeech 1997
ISCA Archive Eurospeech 1997

Language identification with language-independent acoustic models

Cristobal Corredor-Ardoy, Jean Luc Gauvain, Martine Adda-Decker, Lori Lamel

In this paper we explore the use of language- independent acoustic models for language identification (LID). The phone sequence output by a single language-independent phone recognizer is rescored with language-dependent phonotactic models approximated by phone bigrams. The language-independent phoneme inventory was obtained by Agglomerative Hierarchical Clustering, using a measure of similarity between phones. This system is compared with a parallel language-dependent phone architecture, which uses optimally the acoustic log likelihood and the phonotactic score for language identiffication. Experiments were carried out on the 4-language telephone speech corpus IDEAL, containing calls in British English, Spanish, French and German. Results show that the language-independent approach performs as well as the language-dependent one: 9% versus 10% of error rate on 10 second chunks, for the 4-language task.

doi: 10.21437/Eurospeech.1997-41

Cite as: Corredor-Ardoy, C., Gauvain, J.L., Adda-Decker, M., Lamel, L. (1997) Language identification with language-independent acoustic models. Proc. 5th European Conference on Speech Communication and Technology (Eurospeech 1997), 55-58, doi: 10.21437/Eurospeech.1997-41

  author={Cristobal Corredor-Ardoy and Jean Luc Gauvain and Martine Adda-Decker and Lori Lamel},
  title={{Language identification with language-independent acoustic models}},
  booktitle={Proc. 5th European Conference on Speech Communication and Technology (Eurospeech 1997)},