David Klusáček

Home office
office hours
contact me to arrange a meeting
+420 951 554 xxx
+420 257 223 293
Malostranské náměstí 25
118 00 Praha 1
Czech Republic

Main Research Interests

Speech recognition (mainly the acoustic front-end), speaker identification, digital signal processing, functional and logic programming, compilation and optimization techniques, formal program verification, mathematical logic, automated theorem proving, knowledge representation, computational neuroscience and neuroanatomy.

Curriculum Vitae

1991--1995 Electronics secondary school V uzlabine 320, Praha 10. Finished with the highest honors.
1995--2003 Computer Science at the Faculty of Mathematics and Physics, Charles University Prague (MFF-UK). Defended diploma thesis {\sl ``Supercompilation''} and graduated with the highest honors. Master degree.
2003--2012 PhD studies at MFF-UK (UFAL department). Defended doctoral thesis {\sl ``New Methods in Statistical Speech Recognition''}.
Stays 2002 Participated at the WS'02 Johns Hopkins University summer workshop in the {\it ``SuperSID''} group.
2004 Co-op/Intern at IBM T.J. Watson Research Center in Yorktown Heights, New York, doing research concerning speaker detection.
1995 The first prize in national competition of high-school scientific projects in categories {\sl mathematics} (for the theory and implementation of the Fast Hartley Transform) and {\sl electronics} (for the design and manufacturing of PC-compatible sound card).
1996 Czech Ministry of Education Award ``Mathematical Talent 1995'' for the work {\it ``Fast Hartley Transform''}.
2002 One-year research funding won at the end of the JHU workshop (out of 9 competitors) for {\it ``Pronunciation Model''\/}-based speaker detection.


List of classesNPFL109 Digital Sound Processing

Selected Bibliography

Klusacek David: mmi_clustering. Data/software, UFAL https://ufal.mff.cuni.cz/tools/mmic

Navratil Jiri, Klusacek David: On Linear DETs. In: it Acoustics, Speech and Signal Processing, 2007. ICASSP 2007. IEEE International Conference}, Copyright Institute of Electrical and Electronics Engineers (IEEE), Honolulu, HI, USA, ISBN 1-4244-0728-1, ISSN 1520-6149, pp. 229-232, 2007

Klusacek David: Optimal Detection in Case of the Sparse Training Data. In: Proceedings of ODYSSEY04, pp. 97--104, 2004

D. Reynolds, W. Andrews, J. Campbell, J. Navrðtil, D. Klusðñek, B. Peskin, A. Adami, Q. Jin, J. Abramson, R. Michaesku, J. Godfrey and B. Xiang: The SuperSID Project: Exploiting High-level Information for High-accuracy Speaker Recognition. In: Proceedings of ICASSP 2003

D. Klusacek, J. Navratil, D. Reynolds, J. Campbell: Conditional Pronunciation Modeling in Speaker Detection. In: Proceedings of ICASSP 2003