Difference between revisions of "Publications:Fusing Various Audio Feature Sets for Detection of Parkinson's Disease from Sustained Voice and Speech Recordings"
(Created page with "<div style='display: none'> == Do not edit this section == </div> {{PublicationSetupTemplate|PID=955931 |Name=Vaiciukynas, Evaldas (Kaunas University of Technology, Kaunas, Li...") |
|||
| Line 2: | Line 2: | ||
== Do not edit this section == | == Do not edit this section == | ||
</div> | </div> | ||
| − | {{PublicationSetupTemplate|PID=955931 | + | {{PublicationSetupTemplate|Author=Evaldas Vaiciukynas, Antanas Verikas, Adas Gelzinis, Marija Bacauskiene, Kestutis Vaskevicius, Virgilijus Uloza, Evaldas Padervinskis, Jolita Ciceliene |
| − | |Name=Vaiciukynas, Evaldas (Kaunas University of Technology, Kaunas, Lithuania);Verikas, Antanas | + | |PID=955931 |
| + | |Name=Vaiciukynas, Evaldas (Kaunas University of Technology, Kaunas, Lithuania);Verikas, Antanas (av) (0000-0003-2185-8973) (Högskolan i Halmstad (2804), Akademin för informationsteknologi (16904), Halmstad Embedded and Intelligent Systems Research (EIS) (3938), CAISR Centrum för tillämpade intelligenta system (IS-lab) (13650)) (Kaunas University of Technology, Kaunas, Lithuania);Gelzinis, Adas (Kaunas University of Technology, Kaunas, Lithuania);Bacauskiene, Marija (Kaunas University of Technology, Kaunas, Lithuania);Vaskevicius, Kestutis (Kaunas University of Technology, Kaunas, Lithuania);Uloza, Virgilijus (Lithuanian University of Health Sciences, Kaunas, Lithuania);Padervinskis, Evaldas (Lithuanian University of Health Sciences, Kaunas, Lithuania);Ciceliene, Jolita (Lithuanian University of Health Sciences, Kaunas, Lithuania) | ||
|Title=Fusing Various Audio Feature Sets for Detection of Parkinson’s Disease from Sustained Voice and Speech Recordings | |Title=Fusing Various Audio Feature Sets for Detection of Parkinson’s Disease from Sustained Voice and Speech Recordings | ||
|PublicationType=Journal Paper | |PublicationType=Journal Paper | ||
Latest revision as of 22:41, 30 September 2016
Property "Publisher" has a restricted application area and cannot be used as annotation property by a user. Property "Author" has a restricted application area and cannot be used as annotation property by a user. Property "Author" has a restricted application area and cannot be used as annotation property by a user. Property "Author" has a restricted application area and cannot be used as annotation property by a user. Property "Author" has a restricted application area and cannot be used as annotation property by a user. Property "Author" has a restricted application area and cannot be used as annotation property by a user. Property "Author" has a restricted application area and cannot be used as annotation property by a user. Property "Author" has a restricted application area and cannot be used as annotation property by a user. Property "Author" has a restricted application area and cannot be used as annotation property by a user.
| Title | Fusing Various Audio Feature Sets for Detection of Parkinson’s Disease from Sustained Voice and Speech Recordings |
|---|---|
| Author | |
| Year | 2016 |
| PublicationType | Journal Paper |
| Journal | Lecture Notes in Computer Science |
| HostPublication | |
| Conference | 18th International Conference, SPECOM 2016, Budapest, Hungary, August 23-27, 2016 |
| DOI | http://dx.doi.org/10.1007/978-3-319-43958-7_39 |
| Diva url | http://hh.diva-portal.org/smash/record.jsf?searchId=1&pid=diva2:955931 |
| Abstract | The aim of this study is the analysis of voice and speech recordings for the task of Parkinson’s disease detection. Voice modality corresponds to sustained phonation /a/ and speech modality to a short sentence in Lithuanian language. Diverse information from recordings is extracted by 22 well-known audio feature sets. Random forest is used as a learner, both for individual feature sets and for decision-level fusion. Essentia descriptors were found as the best individual feature set, achieving equal error rate of 16.3 % for voice and 13.3 % for speech. Fusion of feature sets and modalities improved detection and achieved equal error rate of 10.8 %. Variable importance in fusion revealed speech modality as more important than voice. © Springer International Publishing Switzerland 2016 |
