13.07.2015 Views

Proceedings Fonetik 2009 - Institutionen för lingvistik

Proceedings Fonetik 2009 - Institutionen för lingvistik

Proceedings Fonetik 2009 - Institutionen för lingvistik

SHOW MORE
SHOW LESS

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

<strong>Proceedings</strong>, FONETIK <strong>2009</strong>, Dept. of Linguistics, Stockholm UniversityFigure 3. The delta SRT value (with SynFace-Without SynFace) per subject with babble noise.Left: the Swedish moderate hearing impaired group. Right: the Dutch cochlear implants subjects.Figure 4. The delta SRT at the 20 item in the listand the 10 th for 40 SRT measurements.DiscussionOverall, the preliminary analysis of the resultsof both the SRT test and the effort scalingshowed limited beneficial effects for SynFace.However, the Swedish participants showed anoverall beneficial effect for the use of SynFacein the SRT test when listening to speech withbabble noise.Another possible approach when examiningthe benefit of using SynFace may be looking atindividual results as opposed to group means.The data shows that some people benefit fromthe exposure to SynFace. In the ongoing analysisof the tests, we will try to see if there arecorrelations in the results for different tests persubject, and hence to study if there are certainfeature which characterize subjects who showconsistent benefit from SynFace throughout allthe tests.ConclusionsThe paper reports on the methods used forthe large scale hearing impaired tests with Syn-Face lip-synchronized talking head. Preliminaryanalysis of the results from the user studieswith hearing impaired subjects where performedat three sites. Although SynFaceshowed consistent advantage for the normalhearing subjects, SynFace did not show a consistentadvantage with the hearing impairedsubjects, but there were SynFace benefits forsome of the subjects in all the tests, especiallyfor speech-in-babble-noise condition.AcknowledgementsThis work has been carried out under theHearing at Home (HaH) project. HaH is fundedby the EU (IST-045089). We would like tothank other project members at KTH, Sweden;HörTech, OFFIS, and ProSyst, Germany;VIATAAL, the Netherlands, and TelefonicaI&D, Spain.ReferencesAgelfors, E., Beskow, J., Dahlquist, M.,Granström, B., Lundeberg, M., Spens,K-E., & Öhman, T. (1998). Syntheticfaces as a lipreading support. In <strong>Proceedings</strong>of ICSLP'98.Agelfors, E., Beskow, J., Karlsson, I., Kewley,J., Salvi, G., & Thomas, N. (2006).User Evaluation of the SYNFACETalking Head Telephone. Lecture Notesin Computer Science, 4061, 579-586.Beskow, J., Granström, B., Nordqvist, P.,Al Moubayed, S., Salvi, G., Herzke, T.,& Schulz, A. (2008). Hearing at Home– Communication support in home environmentsfor hearing impaired persons.In <strong>Proceedings</strong> of Interspeech.Brisbane, Australia.Hagerman, B., & Kinnefors, C. (1995). Efficientadaptive methods for measuringspeech reception threshold in quiet andin noise. Scand Audiol, 24, 71-77.Lindberg, B., Johansen, F. T., Warakagoda,N., Lehtinen, G., Kai, Z., Gank, A.,Elenius, K., & Salvi, G. (2000). A noiserobust multilingual reference recogniserbased on SpeechDat(II). In Proc ofICSLP 2000, (pp. 370-373). Beijing.Siciliano, C., Faulkner, A., & Williams, G.(2003). Lipreadability of a synthetictalking face in normal hearing and hearingimpairedlisteners. In AVSP 2003-International Conference on Audio-Visual Speech Processing.143

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!