Speech emotion recognition for SROL database using weighted KNN algorithm

Author

Feraru, Monica ; Zbancioc, Marius

Author_Institution

Inst. of Comput. Sci., Iasi, Romania

fYear

2013

Firstpage

1

Lastpage

4

Abstract

In this study, we utilized an improved version of the classical KNN algorithm which associates to each parameter from the features vectors weights according to their performance in the classification process. We obtained the recognition percents of emotions around 65-67%, for the Romanian language, on the SROL database, which are comparable with the results for other languages, with non-professional voice database. This is the first study when the parameters are extracted on the sentence level. Until now, the analysis was made on the phoneme level.

Keywords

audio databases; emotion recognition; natural language processing; signal classification; speech recognition; Romanian language; SROL database; classification process; features vector; nonprofessional voice database; phoneme level; sentence level; speech emotion recognition; weighted KNN algorithm; Accuracy; Classification algorithms; Databases; Emotion recognition; Feature extraction; Speech; Speech recognition; emotion recognition; prosodic features; weighted KNN;

fLanguage

English

Publisher

ieee

Conference_Titel

Electronics, Computers and Artificial Intelligence (ECAI), 2013 International Conference on

Conference_Location

Pitesti

Print_ISBN

978-1-4673-4935-2

Type

conf

DOI

10.1109/ECAI.2013.6636198

Filename

6636198