Prezentace,el.dokumenty

 
Černocký, J., Szőke, I., Hannemann, M., Kombrink, S.: Word-subword based keyword spotting with implications in OOV detection, Pacific Grove, US, IEEE, 2010, s. 34
Jazyk publikace:angličtina
Název publikace:Word-subword based keyword spotting with implications in OOV detection
Název (cs):Slovní a podslovní detekce klíčových slov a její implikace v detekci OOV
Strany:34
Konference:Asilomar Conference on Signals, Systems, and Computers
Místo vydání:Pacific Grove, US
Rok:2010
Vydavatel:Institute of Electrical and Electronics Engineers
URL:http://www.fit.vutbr.cz/research/groups/speech/publi/2010/asilomar_kwd_oov.ppt [PPT]
Klíčová slova
speech recognition, keyword spotting, spoken term detection, OOV
Anotace
Práce je o hybridní slovní a podslovní detekci klíčových slov. Tato detekce zachovává výhody LVCRS a dovoluje hledat slova mimo slovník (OOV) jako sekvence pod-slovních jednotek.
Abstrakt
Main-stream systems for keyword spotting and spoken term detection are based on the series of Large Vocabulary Continuous Speech Recognizer with subsequent search in its output. These systems are limited by the vocabulary of the recognizer and are not able to detect Out of Vocabulary (OOV) words. This talk will present our work in designing hybrid word-subword keyword spotting systems, that maintain the accuracy of LVCSR, while allowing for detecting OOVs as sequences of sub-word units. We will also show the links of this work to the detection, description and clustering of OOVs, as investigated in the framework of the EC-sponsored project DIRAC.