BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Speech Synthesis at Google - Matt Stuttle\, Google
DTSTART:20120118T141500Z
DTEND:20120118T151500Z
UID:TALK35382@talks.cam.ac.uk
CONTACT:Stephen Clark
DESCRIPTION:This talk will give an overview of the basics of speech synthe
 sis - from\nthe front-end text-normalisation issues of cleaning up ambiguo
 us text input\nto the waveform-synthesis backend task of turning a linguis
 tic\nspecification into speech.  The two main techniques of unit\nselectio
 n/concatenation and parametric HMM  approaches will be covered\,\nwith the
  advantages of each.\n\nThe second part of the talk with focus on the appl
 ications and rollout of\nthese speech synthesis technologies at Google\, d
 ealing with requests from\ndozens of product teams and balancing requireme
 nts and challenges for the\nfuture and scaling systems for embedded and se
 rver applications and the\nexperience of an acquired startup technology te
 am at Google.\n\nMatt Stuttle is a Research Scientist at Google working in
  the Speech Synthesis team in London.  Matt was VP Research at Phonetic A
 rts\, a startup focused on speech synthesis technologies for computer game
 s\nand games development\, which was acquired by Google in 2010.  Prior t
 o working at Phonetic Arts\, he was at Toshiba's Cambridge Research Lab do
 ing research and development of speech recognition and synthesis\ntechnolo
 gies including speech-to-speech translation and in-car and mobile based te
 chnologies.  Matt holds a PhD from the Machine Intelligence lab at CUED i
 n speech recognition and was also a Research\nAssociate in statistical s
 poken dialog systems in the same group.\n
LOCATION:Lecture Theatre 1\, Computer Laboratory
END:VEVENT
END:VCALENDAR