BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Drupal iCal API//EN
X-WR-CALNAME:Events items teaser
X-WR-TIMEZONE:America/Toronto
BEGIN:VTIMEZONE
TZID:America/Toronto
X-LIC-LOCATION:America/Toronto
BEGIN:DAYLIGHT
TZNAME:EDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
DTSTART:20190310T070000
END:DAYLIGHT
BEGIN:STANDARD
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
DTSTART:20191103T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:69b15aa214c31
DTSTART;TZID=America/Toronto:20191129T100000
SEQUENCE:0
TRANSP:TRANSPARENT
DTEND;TZID=America/Toronto:20191129T100000
URL:https://uwaterloo.ca/artificial-intelligence-group/events/masters-thesi
 s-presentation-honkling-browser-personalization
LOCATION:DC - William G. Davis Computer Research Centre 200 University Aven
 ue West 2310 Waterloo ON N2L 3G1 Canada
SUMMARY:Master’s Thesis Presentation: Honkling: In-Browser Personalizatio
 n\nfor Ubiquitous Keyword Spotting
CLASS:PUBLIC
DESCRIPTION:JAEJUN LEE\, MASTER’S CANDIDATE\n_David R. Cheriton School of
  Computer Science_\n\nUsed for simple voice commands and wake-word detecti
 on\, keyword\nspotting (KWS) is the task of detecting pre-determined keywo
 rds in a\nstream of utterances. A common implementation of KWS involves\nt
 ransmitting audio samples over the network and detecting target\nkeywords 
 in the cloud with neural networks because on-device\napplication developme
 nt presents compatibility issues with various\nedge devices and provides l
 imited supports for deep learning.\nUnfortunately\, such an architecture c
 an lead to unpleasant user\nexperience because network latency is not dete
 rministic. Furthermore\,\nthe client-server architecture raises privacy co
 ncerns because users\nlose control over the audio data once it leaves the 
 edge device. 
DTSTAMP:20260311T120554Z
END:VEVENT
END:VCALENDAR