CONF Motlicek_INTERSPEECH2009-2_2009/IDIAP Automatic Out-of-Language Detection Based on Confidence Measures Derived fromLVCSR Word and Phone Lattices Motlicek, Petr EXTERNAL http://publications.idiap.ch/attachments/papers/2009/Motlicek_INTERSPEECH2009-2_2009.pdf PUBLIC ISCA - 10thAnnual Conference of the International Speech Communication Association Brighton, England 2009 ISCA 2009 September 2009 1215-1218 1990-9772 Confidence Measures (CMs) estimated from Large Vocabulary Continuous Speech Recognition (LVCSR) outputs are commonly used metrics to detect incorrectly recognized words. In this paper, we propose to exploit CMs derived from frame-based word and phone posteriors to detect speech segments containing pronunciations from non-target (alien) languages. The LVCSR system used is built for English, which is the target language, with medium-size recognition vocabulary (5k words). The efficiency of detection is tested on a set comprising speech from three different languages (English, German, Czech). Results achieved indicate that employment of specific temporal context (integrated in the word or phone level) significantly increases the detection accuracies. Furthermore, we show that combination of several CMs can also improve the efficiency of detection.