Automatic detection and evaluation of edentulous speakers with insufficient dentures
Standard
Automatic detection and evaluation of edentulous speakers with insufficient dentures. / Bocklet, Tobias; Hönig, Florian; Haderlein, Tino; Stelzle, Florian; Knipfer, Christian; Nöth, Elmar.
In: LECT NOTES COMPUT SC, 2010, p. 243-250.Research output: SCORING: Contribution to journal › Conference article in journal › Research › peer-review
Harvard
APA
Vancouver
Bibtex
}
RIS
TY - JOUR
T1 - Automatic detection and evaluation of edentulous speakers with insufficient dentures
AU - Bocklet, Tobias
AU - Hönig, Florian
AU - Haderlein, Tino
AU - Stelzle, Florian
AU - Knipfer, Christian
AU - Nöth, Elmar
N1 - Copyright: Copyright 2010 Elsevier B.V., All rights reserved.
PY - 2010
Y1 - 2010
N2 - Dental rehabilitation by complete dentures is a state-of-the-art approach to improve functional aspects of the oral cavity of edentulous patients. It is important to assure that these dentures have a sufficient fit. We introduce a dataset of 13 edentulous patients that have been recorded with and without complete dentures in situ. These patients have been rated an insufficient fit of their dentures, so that additional (sufficient) dentures and additional speech recordings have been prepared. In this paper we show that sufficient dentures increase the performance of an ASR system by ca. 27 %. Based on these results, we present and discuss three different systems that automatically determine whether the dentures of an edentulous person have a sufficient fit or not. The system with the best performance models the recordings by GMMs and uses the mean vectors of these GMMs as features in an SVM. With this system we were able to achieve a recognition rate of 80 %.
AB - Dental rehabilitation by complete dentures is a state-of-the-art approach to improve functional aspects of the oral cavity of edentulous patients. It is important to assure that these dentures have a sufficient fit. We introduce a dataset of 13 edentulous patients that have been recorded with and without complete dentures in situ. These patients have been rated an insufficient fit of their dentures, so that additional (sufficient) dentures and additional speech recordings have been prepared. In this paper we show that sufficient dentures increase the performance of an ASR system by ca. 27 %. Based on these results, we present and discuss three different systems that automatically determine whether the dentures of an edentulous person have a sufficient fit or not. The system with the best performance models the recordings by GMMs and uses the mean vectors of these GMMs as features in an SVM. With this system we were able to achieve a recognition rate of 80 %.
KW - applied system
KW - assistive technology
KW - speech recognition
KW - user modeling
UR - http://www.scopus.com/inward/record.url?scp=78049295631&partnerID=8YFLogxK
U2 - 10.1007/978-3-642-15760-8_31
DO - 10.1007/978-3-642-15760-8_31
M3 - Conference article in journal
AN - SCOPUS:78049295631
SP - 243
EP - 250
JO - LECT NOTES COMPUT SC
JF - LECT NOTES COMPUT SC
SN - 0302-9743
T2 - 13th International Conference on Text, Speech and Dialogue, TSD 2010
Y2 - 6 September 2010 through 10 September 2010
ER -