BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Automated Measures for Assessment of Prosody in Autism - Emily Tuc
 ker Prud'hommeaux\, Oregon Health and Science University
DTSTART:20091109T190000Z
DTEND:20091109T200000Z
UID:TALK21354@talks.cam.ac.uk
CONTACT:Bert Vaux
DESCRIPTION:Assessment of prosody is important for diagnosis and\nremediat
 ion of speech and language disorders\, for foreign language instruction\, 
 and for diagnosis of neurological conditions\, including Autism Spectrum D
 isorder (ASD).  Current assessment methods are\nlargely auditory-perceptua
 l. In this talk\, I propose an alternative approach for evaluating focus a
 nd lexical stress automatically using speech technology.  Speech data were
  collected from children with ASD and with typical development (TD) who pe
 rformed three tasks\, each\ndesigned to elicit a contrast in stress or foc
 us.  Our automated analysis of these data combined global spectral\, funda
 mental frequency\, and temporal information into a single score. These aut
 omatically computed scores were validated against scores both from\nnaïve
  judges and trained clinicians.  We found that these automated measures co
 rrelated with the scores produced by the judges better than the clinicians
 ’ scores did and as well as the individual judges\ncorrelated with one a
 nother.  The automatically generated scores also differentiated significan
 tly between the ASD and TD groups but revealed that the acoustic features 
 that predict auditory-perceptual\njudgment are not the same as those that 
 distinguish the two diagnostic groups.
LOCATION:Beves Room\, King's College\, Cambridge
END:VEVENT
END:VCALENDAR