CONF Motlicek_INTERSPEECH2010_2010/IDIAP English Spoken Term Detection in Multilingual Recordings Motlicek, Petr Valente, Fabio Garner, Philip N. Confidence Measure (CM) LVCSR Out-Of-Language (OOL) detection Spoken Term Detection (STD) https://publications.idiap.ch/index.php/publications/showcite/Motlicek_Idiap-RR-21-2010 Related documents ISCA - Proceedings of Interspeech, Makuhari, Japan, 2010 Makuhari, Japan 2010 September 2010 This paper investigates the automatic detection of English spoken terms in a multi-language scenario over real lecture recordings. Spoken Term Detection (STD) is based on an LVCSR where the output is represented in the form of word lattices. The lattices are then used to search the required terms. Processed lectures are mainly composed of English, French and Italian recordings where the language can also change within one recording. Therefore, the English STD system uses an Out-Of-Language (OOL) detection module to filter out non-English input segments. OOL detection is evaluated w.r.t. various confidence measures estimated from word lattices. Experimental studies of OOL detection followed by English STD are performed on several hours of multilingual recordings. Significant improvement of OOL+STD over a stand-alone STD system is achieved (relatively more than 50% in EER). Finally, an additional modality (text slides in the form of PowerPoint presentations) is exploited to improve STD. REPORT Motlicek_Idiap-RR-21-2010/IDIAP English Spoken Term Detection in Multilingual Recordings Motlicek, Petr Valente, Fabio Garner, Philip N. https://publications.idiap.ch/index.php/publications/showcite/Motlicek_INTERSPEECH2010_2010 Related documents Idiap-RR-21-2010 2010 Idiap Rue Marconi 19, Martigny, 1920, Switzerland July 2010 This paper investigates the automatic detection of English spoken terms in a multi-language scenario over real lecture recordings. Spoken Term Detection (STD) is based on an LVCSR where the output is represented in the form of word lattices. The lattices are then used to search the required terms. Processed lectures are mainly composed of English, French and Italian recordings where the language can also change within one recording. Therefore, the English STD system uses an Out-Of-Language (OOL) detection module to filter out non-English input segments. OOL detection is evaluated w.r.t. various confidence measures estimated from word lattices. Experimental studies of OOL detection followed by English STD are performed on several hours of multilingual recordings. Significant improvement of OOL+STD over a stand-alone STD system is achieved (relatively more than 50% in EER). Finally, an additional modality (text slides in the form of PowerPoint presentations) is exploited to improve STD.