BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Poly AI: Architecting Human-Sounding Voice Assistants - Razvan Kus
 ztos 
DTSTART:20220302T130500Z
DTEND:20220302T135500Z
UID:TALK169742@talks.cam.ac.uk
CONTACT:Ben Karniely
DESCRIPTION:Voice assistants pose a series of technical challenges such as
  speech recognition\, language understanding\, and much more. However\, as
  PolyAI rides its commercial journey\, we increasingly realise the ability
  to customise and give back control to the designer is paramount. \n\nIn t
 his talk\, I want to share how this vision of customisability shapes our t
 echnical decisions\, from our runtime infrastructure\, multi-tenant bot bu
 ilding studio\, to how we structure our machine learning pipelines\, as we
 ll as the less well known world of telephony stack. Through this design pr
 inciple\, we were able to share a new generation of voice assistants to mi
 llions of users. Hope to see you there!\n\nPlease sign-up at the following
  link by Friday 25th February if you would like to attend:\nhttps://forms.
 gle/8Z7eDWE5AwR5VLA87\n\nThe speaker is an alumnus of the Department.\n\nS
 ome catering will be provided outside the lecture theatre following the ta
 lk.\n\nWe are asking those attending in person to please take a lateral fl
 ow test with a negative result before the talk.\n
LOCATION:Lecture Theatre 1\, Computer Laboratory\, William Gates Building.
END:VEVENT
END:VCALENDAR