BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:How can speech technologies support learners to improve their skil
 ls of speaking\, listen-ing\, conversation and more? - Nobuaki MINEMATSU 
DTSTART:20190319T140000Z
DTEND:20190319T153000Z
UID:TALK121861@talks.cam.ac.uk
CONTACT:Anton Ragni
DESCRIPTION:In the globalization era\, not only students but also immigran
 t workers have to learn new languages for smooth oral communication in tho
 se languages. In this talk\, the lecturer illustrates how speech technolog
 ies\, i.e. speech synthesis\, speech recognition\, voice conversion\, etc 
 can support learners to improve their skills of speaking\, listening\, con
 versation\, and more. Text does not show any prosodic structure explicitly
  and native speakers use their implicit knowledge on prosodic control to r
 ead aloud that text naturally. Implicit knowledge is difficult for teacher
 s to explain explicitly and therefore prosody training is rare in classroo
 ms. Text-to-speech systems often use a text-based prosody prediction modul
 e and this module is used effectively to teach prosodic control required t
 o read given texts aloud explicitly to learners. In High Variability Phone
 tic Training (HVPT)\, teachers use speech stimuli with different ages\, ge
 nders\, accents\, background noises\, etc. Being exposed to those variabil
 ities\, learners can obtain robust listening skills. However\, teachers pr
 epare those stimuli manually. By introducing speech analysis and voice con
 version techniques\, those variabilities are easily enhanced. In the talk\
 , an interesting example of adversarial training\, which was originally us
 ed for machine learners and is newly introduced to human learners\, and it
 s effectiveness for acquiring robust listening skills are explained. Furth
 er\, use of speech recognition technologies for shadowing assessment to im
 prove parallel processing skills for conversation is described. In the lec
 turer’s laboratory\, a new project has started to realize a novel speech
  assessment framework\, where not native-likeness but comprehensibility of
  learners’ speech is mainly focused on for assessment. The lecturer show
 s recently obtained results of objective measurement of comprehensibility 
 of learners’ speech.
LOCATION:Department of Engineering - Lecture Theater 6
END:VEVENT
END:VCALENDAR