ISCA Archive Interspeech 2005
ISCA Archive Interspeech 2005

Fully automated system for Czech spoken broadcast transcription with very large (300k+) lexicon

Jan Nouza, Jindrich Zdánský, Petr David, Petr Cerva, Jan Kolorenc, Dana Nejedlová

We present a system developed for fully automated processing of Czech spoken broadcast programs. It includes modules for unsupervised segmentation of audio stream, speaker and gender recognition followed by speaker adaptation, and own speech decoder designed for extremely large vocabularies. Compared to our previous results reported in 2004, the new system reduced the WER (evaluated on the Czech part of the European COST Broadcast News Database) from 28.5% to 18.4%. This significant improvement was accomplished namely due to the larger lexicon (312K) with multiple text and pronunciation variants and multi-word entries, speaker and gender adapted acoustic matching and improved language modeling. Besides the results achieved in the Broadcast News task we refer also about the performance in other similar jobs, like the transcription of a talk show or parliament speech.


doi: 10.21437/Interspeech.2005-548

Cite as: Nouza, J., Zdánský, J., David, P., Cerva, P., Kolorenc, J., Nejedlová, D. (2005) Fully automated system for Czech spoken broadcast transcription with very large (300k+) lexicon. Proc. Interspeech 2005, 1681-1684, doi: 10.21437/Interspeech.2005-548

@inproceedings{nouza05_interspeech,
  author={Jan Nouza and Jindrich Zdánský and Petr David and Petr Cerva and Jan Kolorenc and Dana Nejedlová},
  title={{Fully automated system for Czech spoken broadcast transcription with very large (300k+) lexicon}},
  year=2005,
  booktitle={Proc. Interspeech 2005},
  pages={1681--1684},
  doi={10.21437/Interspeech.2005-548}
}