Faculty of Information Technology, BUT

Publication Details

Posterior-based Out of Vocabulary Word Detection in Telephone Speech

KOMBRINK Stefan, BURGET Lukáš, MATĚJKA Pavel, KARAFIÁT Martin and HEŘMANSKÝ Hynek. Posterior-based Out of Vocabulary Word Detection in Telephone Speech. In: Proc. Interspeech 2009. Brighton: International Speech Communication Association, 2009, pp. 80-83. ISSN 1990-9772.
Czech title
Detekce slov mimo slovník v telefonní řeči založená na posteriorech
Type
conference paper
Language
english
Authors
URL
Keywords
confidence measures, out-of-vocabulary word detection, phone posteriors, neural net, OOV
Abstract
The paper is on Posterior-based Out of Vocabulary Word Detection in Telephone Speech
Annotation
In this paper we present an out-of-vocabulary word detector suitable for English conversational and read speech. We use an approach based on phone posteriors created by a Large Vocabulary Continuous Speech Recognition system and an additional phone recognizer, that allows detection of OOV and misrecognized words. In addition, the recognized word output can be transcribed more detailed using several classes. Reported results are on CallHome English and Wall Street Journal data.
Published
2009
Pages
80-83
Journal
Proceedings of Interspeech, no. 9, ISSN 1990-9772
Proceedings
Proc. Interspeech 2009
Conference
Interspeech 2009, Brighton, GB
Publisher
International Speech Communication Association
Place
Brighton, GB
BibTeX
@INPROCEEDINGS{FITPUB9037,
   author = "Stefan Kombrink and Luk\'{a}\v{s} Burget and Pavel Mat\v{e}jka and Martin Karafi\'{a}t and Hynek He\v{r}mansk\'{y}",
   title = "Posterior-based Out of Vocabulary Word Detection in Telephone Speech",
   pages = "80--83",
   booktitle = "Proc. Interspeech 2009",
   journal = "Proceedings of Interspeech",
   number = 9,
   year = 2009,
   location = "Brighton, GB",
   publisher = "International Speech Communication Association",
   ISSN = "1990-9772",
   language = "english",
   url = "https://www.fit.vut.cz/research/publication/9037"
}
Back to top