Ondřej Dušek

office
N233
office hours
email me to arrange a meeting
email
odusek@ufal.mff.cuni.cz
address
IMPAKT – „N“
V Holešovičkách 747/2
180 00 Praha 8
Czech Republic

Main Research Interests

  • Natural language generation
  • Spoken dialogue systems and chatbots
  • Machine learning
  • Machine translation

Projects

Current

  • THEaiTRE – Automatically generating a theatre play (Czech Technical Agency, 2020-2022)
  • NaMuDDiS – Natural Multi-domain Dialogue Systems (Charles University funded, 2019-2021)
  • METOD – dialogue management (industry cooperation with Agnostix co-funded by City of Prague, 2020)

Past

At Heriot-Watt University in Edinburgh (2016-2018):

During my Ph.D. study (2011-2016):

  • AdaNLG – adaptive natural language generator (2014-2016)
  • Vystadial – statistical spoken dialogue system (2013-2016)
  • QTLeap – semantic machine translation (2013-2016)
  • Khresmoi – medical information retrieval (working on machine translation, 2013-2014)
  • FAUST – improving machine translation fluency (2011-2013)

Teaching

List of classesNPFL099 Statistical dialogue systemsNPFL123 Dialogue systems

Students

Students I co-supervise and co-supervised:

News

2020/07/25: We got two papers on neural text-to-speech accepted to Interspeech, both are available on arXiv: Tomáš Nekvinda's One Model, Many Languages: Meta-learning for Multilingual Text-to-Speech and Jan Vainer's SpeedySpeech: Efficient Neural Speech Synthesis.

2020/07/01: Me & Vojtěch started working on a joint project METOD with Agnostix, investigating dialogue management modelling and visualizations, supported by the City of Prague.

2020/05/26: We (Jindřich Libovický, Zdeněk Kasner, Jindřich Helcl & me) took part in the STAPLE/Duolingo translation & paraphrasing WNGT shared task. Our report is now in the ACL anthology.

2020/04/06: Our paper “Fact-based Content Weighting for Evaluating Abstractive Summarisation” (by Xinnuo Xu, me, Jingiy Li, Yannis Konstas & Verena Rieser) was accepted for ACL. You can read the paper in the ACL anthology.

2020/04/01: Our grant project THEaiTRE for automatically generating theatre plays was accepted by the Czech Technical Agency (and actually got the highest ranking of all submissions)! We plan to have a theatre play performed in January 2021, for the 100th anniversary of Karel Čapek's R.U.R.

2020/03/23: Verena Rieser and I were guests at the NLP Highlights podcast, and we talked about data-to-text generation. You can listen to the podcast on Soundcloud.

2020/03/03: I gave a short talk at the Czechbots chatbot meetup. You can find the slides here.

2020/02/18: Our undergrad Dialogue systems course started its 2020 run – lecture slides will appear on the course webpage.

2019/10/31: I presented 3 papers at INLG 2019: Neural Generation for Czech: Data and Baselines (work with Filip Jurčíček), Semantic Noise Matters for Neural Natural Language Generation (work with Dave Howcroft and Verena Rieser), and Automatic Quality Estimation for Natural Language Generation: Ranting (Jointly Rating and Ranking) (work with Karin Sevegnani, Ioannis Konstas and Verena Rieser).

2019/10/03: We started teaching a new course on Statistical dialogue systems – lecture slides will appear on the course webpage.

2019/09/13: Our paper User evaluation of a multi-dimensional statistical dialogue system (with Simon Keizer, Xingkun Liu and Verena Rieser) was presented at SIGDIAL in Stockholm and is now online on arXiv.

2019/07/10: Our paper analyzing the E2E NLG Challenge results is now published in Computer Speech and Language and available online as open access.

Biographical

Tools I'm participating on

  • Alex – spoken dialogue system framework
  • Flect – statistical morphology generation
  • MTMonkey – machine translation web services infrastructure
  • RatPred – trainable NLG quality estimation
  • TGen – a statistical natural language generator
  • Treex – a modular NLP toolkit

Personal webpage