BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Investigation of multilingual speech-to-text systems for use in sp
 oken term detection - Kate Knill (University of Cambridge) 
DTSTART:20140403T110000Z
DTEND:20140403T120000Z
UID:TALK50697@talks.cam.ac.uk
CONTACT:Rogier van Dalen
DESCRIPTION:The development of high-performance speech processing systems 
 for low-resource languages is a challenging research area. One approach to
  address the lack of resources is to make use of data from multiple langua
 ges. A popular direction in recent years is to use bottleneck features or 
 hybrid systems trained on multilingual data for speech-to-text (STT) syste
 ms. This talk presents an overview of these approaches for STT\, and their
  performance for both speech recognition and spoken term detection. Experi
 ments will be presented based on the IARPA Babel limited language pack cor
 pora (10 hours/language) using 7 languages for multilingual system develop
 ment and 3 held-out target languages.
LOCATION:Department of Engineering - LR12
END:VEVENT
END:VCALENDAR