This is an archived page of the old course taught by Filip Jurčíček until 2016.
Please see the current course here.


Statistical dialogue systems

The course aims to provide students with basic understanding of methods and approaches to building spoken dialogue systems. The course will emphasize and discuss the importance of statistical techniques in construction of natural and robust dialogue systems. The course is taught in Czech or English. 

During the course, you will have the opportunity to participate in development of the Alex Diaqlogue Systems Framework.

If you are interested in the subject and want to sign up, then you can do it here.


An example of a statistical dialogue system, I worked on while being in Cambridge, UK, can be seen at the following videos: and



  • use of dialogue systems
  • basic components of dialogue systems
  • current approaches to design of dialogue systems
  • knowledge representation in dialogue systems

Spoken language understanding

  • definition of dialogue acts
  • handcrafted methods based on grammars and rules
  • statistical methods

Dialogue management

  • types of dialogue policies
  • representation of a dialogue as MDP or POMDP
  • use of reinforcement learning
  • belief monitoring for dialogue state estimation
  • user modeling
  • user simulation

Natural language generation

  • template-based language generation
  • statistical methods
  • variability and personalization of the generated utterances

Evaluation of dialogue systems

  • testing in laboratory conditions
  • use of crowd-sourcing techniques
  • testing in production environment


[1] J. Psutka, L. Muller, J. Matoušek, and V. Radavá, Mluvíme s počítačem česky. Prague: Academia, 2006, p. 752.

[2] B. Thomson and S. Young, “Bayesian update of dialogue state: A POMDP framework for spoken dialogue systems,” Computer Speech & Language, vol. 24, no. 4, pp. 562-588, 2010.

[3] S. Young et al., “The Hidden Information State model: A practical framework for POMDP-based spoken dialogue management,” Computer Speech & Language, vol. 24, no. 2, pp. 150-174, 2010.

[4] C. M. Bishop, Pattern Recognition and Machine Learning, vol. 4, no. 4. Springer, 2006, p. 738.