TU Berlin

Quality and Usability LabThilo Michael

Page Content

to Navigation

Thilo Michael


Research Field

Quality of speech and Spoken Dialogue Systems


Research Topics

  • Simulation of spoken dialogue
  • Modelling of conversational behaviour (especially turn taking and intelligibility)
  • Prediction of Conversational Quality



Thilo Michael studied Applied Computer Science (B.Sc.) at the Duale Hochschule Baden-Württemberg while working at IBM. In 2016 he finished his Master in Computer Science at the TU Berlin. In March 2017 he joined the Quality and Usability Lab to work on the simulation of human-to-human conversation during impaired speech transmissions.



University politics

I am active in various university committees. I am especially involved in the study programs Medieninformatik & Medientechnik (Media Technology), was active (2017-2019) in the Ausbildungskommission (education commission) and am currently (2019-2021) an elected member of the Fakultätsrat (faculty council). I am member of the WM-Forum.



I am active in teaching since I joined the Quality & Usability Lab in 2017. Since 2020 I work as a researcher with teaching responsibility. I do student projects (almost) every semester covering a wide range of topics (chatbots, spoken dialogue systems, conversation tests, smart home hardware projects, ...). I am responsible for the spoken dialogue exercise in the module Speech Communication and support the exercises of the module Einführung in die Medieninformatik (introduction to media informatics).

I supervise many bachelor and master theses that fall into the topic of my research and related interesting fields. If you are interested in writing your thesis, please feel free to contact me via E-Mail and look for topics on the Jobs & Theses page.



E-Mail: thilo.michael@tu-berlin.de

Phone:   +49 (30) 8353-54252

(I'm often not reachable via phone due to home office)


Postal Address

Quality and Usability Lab

Technische Universität Berlin

Ernst-Reuter-Platz 7

D-10587 Berlin, Germany 






Google Scholar



Quick Access

Schnellnavigation zur Seite über Nummerneingabe