This paper presents the results in designing, processing and evaluation of Serbian emotional speech database. The database has been collected in order to evaluate how well the emotional state in emotional speech is identified by humans and how well the emotional information contained in speech can be used in human-computer communication. The database contains recordings from six actors, three of each gender. Actors were used for the recordings the following emotions: neutral, anger, happiness, sadness and fear. The database consists of: 32 isolated words, 30 short semantically neutral sentences, 30 long semantically neutral sentences and one passage with 79 words in size. A listening test showed the correctly identification of emotions in 95%, and that most confusion occurred between anger and happiness, and between neutral and fear. Finally, the preliminary results in acoustic feature measurements on passage were presented.
Cite as: Jovicic, S.T., Kasic, Z., Dordevic, M., Rajkovic, M. (2004) Serbian emotional speech database: design, processing and evaluation. Proc. 9th Conference on Speech and Computer (SPECOM 2004), 77-81
@inproceedings{jovicic04_specom,
author={Slobodan T. Jovicic and Zorka Kasic and Miodrag Dordevic and Mirjana Rajkovic},
title={{Serbian emotional speech database: design, processing and evaluation}},
year=2004,
booktitle={Proc. 9th Conference on Speech and Computer (SPECOM 2004)},
pages={77--81}
}