Affective computing is a prominent field of research in the domain of data science. The challenging task of affect recognition aims to create an environment of affect-aware computers which acquire a user’s affective state by using complex mathematical models. In order to successfully assess the user’s affective state, aforementioned models should use features that will ensure unique representation of an examined artifact that reveals affective information in the data. The human voice is an attractive modality for applications in the affective computing because of availability and ease of capturing of the affective information. Besides that, demanding cognitive processes and stress leave their mark on speech production which makes speech signals desirable inputs for cognitive load and stress estimation. Since information about one’s affective state, cognitive load or level of stress is hidden in a speech waveform, additional processing of a speech signal is required. The main objective of this Special Session is to discuss different speech analysis techniques with focus on extracting features and employing methods for alternative speech signal representation in order to emphasize artifacts related to affect. Topics of interest include (but are not limited to):

Topics of interest

  • Emotion classification by using deep learning or machine learning methods.
  • Emotion prediction by using deep learning or machine learning methods.
  • Cognitive load estimation.
  • Cognitive load prediction.
  • Stress estimation.
  • Stress prediction.
  • Real time applications of all of the above.


  • Submission deadline: 25 May, 2019
  • Symposium: 23 Sep, 2019 – 25 Sep, 2019


Interested authors are invited to submit papers to this special session through ISPA online submission system by marking the check box on the submission page corresponding to the desired special session. Authors should follow the author instructions for regular ISPA papers. The deadline for online paper submission for this special session is the same as the deadline for regular papers (May 25, 2019).

Papers submitted to this special session will be subject to peer-review. Accepted special session papers will be published in the Symposium Proceedings in the same length and format as regular ISPA papers.


Prof. Davor Petrinović, Ph. D. (
Department of Electronic Systems and Information Processing, University of Zagreb, Faculty of Electrical Engineering and Computing (