About ETRO  |  News  |  Events  |  Vacancies  |  Contact  
Home Research Education Industry Publications About ETRO

Master theses

Current and past ideas and concepts for Master Theses.

Expressive speaking agents


In expressing emotions, humans naturally use various short, affectively charged, non-speech vocal sounds and expressions with some degree of phonemic structure in addition to the speech (e.g. “Yuk!”,”hiii?”, or “rrrrrr”, etc.). These sounds are referred to as affect bursts. There are also sounds, which might be referred to as verbal 'interjections' or ‘fillers’ that are short phonemic structures, heavily charged affectively. Although humans naturally use them in their normal conversations, many currently available voice assistants and text-to-speech engines are not taking full advantage of these sounds. Improving their expressivity would be a way forward to enhance the quality of the interaction with their users.

The main goal of this thesis is to investigate in incorporating affect bursts/interjections into the responses of a speaking agent to be able to improve the expressiveness and thus to enhance the interaction.

Kind of work

The proposed research will consists of the following steps:

  • Analysis of how the affect bursts and interjections are utilized in normal conversations [1],[2]
  • Identifying and categorizing the affect bursts & interjections on an existing expressive speech database [3]
  • Developing an affect burst incorporation methodology for speaking agents to improve their expressiveness
  • Designing a scenario/game with an existing smart speaker (Amazon Echo, Google Home, etc.)/(or with the robot Nao) in which many emotions can be stimulated to evaluate the system in an experiment
  • Thesis report and presentation

Framework of the Thesis

This thesis is related to the speech processing research track at the audiovisual signal processing group at ETRO.

[1] Scherer, K. R. (2014). Affect bursts. In Emotions (pp. 175-208). Psychology Press.
[2] Schröder, M. (2003). Experimental study of affect bursts. Speech communication, 40(1-2), 99-116
[3] Yilmazyildiz, S., Henderickx, D., Vanderborght, B., Verhelst, W., Soetens, E., & Lefeber, D. (2011). EMOGIB: emotional gibberish speech database for affective human-robot interaction. In Affective computing and intelligent interaction (pp. 163-172). Springer, Berlin, Heidelberg

Number of Students

1 or 2

Expected Student Profile

  • Following a MSc in a field related to one or more of the following: applied computer science, computer science and electrical engineering
  • Interest in voice controlled devices and speech signal processing
  • Good programming skills
  • Interest in designing and performing experiments with humans

Prof. Hichem Sahli

+32 (0)2 629 2916

more info


Dr. Selma Yilmazyildiz

+32 (0)2 629 2980

more info


- Contact person




- Contact person

- Thesis proposals

- ETRO Courses

- Contact person

- Spin-offs

- Know How

- Journals

- Conferences

- Books

- Vacancies

- News

- Events

- Press


ETRO Department

Tel: +32 2 629 29 30

©2019 • Vrije Universiteit Brussel • ETRO Dept. • Pleinlaan 2 • 1050 Brussels • Tel: +32 2 629 2930 (secretariat) • Fax: +32 2 629 2883 • WebmasterDisclaimer