Milos Cernak's research pages
 |
Institut of Informatics
Slovak Academy of Sciences
Dubravska cesta 9
845 07 Bratislava, Slovakia
Email: name.surname at ieee.org
|
Short bio
Milos studied Telecommunication engineering at Slovak Technical University in Bratislava, where he obtained a Bachelor degree (1999) and a Master degree (2001) both in Informatics, and a PhD degree (2005) in Telecommunication engineering. During his PhD study, where he was focused on basic and applied Text-to-Speech (TTS) research he spent 3 months in 2002 at Virtual Reality Application Center at Iowa State University in the USA as a Boeing scholar. Milos worked for Slovak Academy of Sciences in 2004-2005, mainly focused on speech analysis,synthesis and recognition techniques. In 2006 Milos moved to France, where he worked with Prof. Chris Wellekens under the aegis of EC 6th Framework project DIVINES, as a post-doc engineer at Institute Eurecom in Sophia-Antipolis. There he was dealing with a diagnostics of error behaviors of machine speech recognizers. In 2007 was Milos a consultant of Acapela R&D TTS group. Now he is again with Institute of Informatics of Slovak Academy of Sciences. Milos is a member of IEEE Signal Processing Society.
Research results and publications
- S. Benus, M. Cernak, M. Rusko, M. Trnka and S. Darjaa
Adapting Slovak ASR for native Germans speaking Slovak. In: Conference on Empirical Methods in Natural Language Processing (EMNLP) 2011, Workshop on Algorithms and Resources for Modelling of Dialects and Language Varieties, DIALECTS 2011, Edinburgh, July 2011, (accepted)
- S. Darjaa, M. Cernak, S. Benus, M. Rusko, R. Sabo and M. Trnka
Rule-Based Triphone Mapping for Acoustic Modeling in Automatic Speech Recognition. In: Text, Speech and Dialogue, Lecture Notes in Computer Science, 2011, (accepted)
- S. Darjaa, M. Cernak, M. Trnka, M. Rusko and R. Sabo
Effective Triphone Mapping for Acoustic Modeling in Speech Recognition. In: Interspeech'11, 12th Annual Conference of the International Speech Communication, August 28-31, Florence, Italy, (accepted)
- M. Cernak
Diagnostics for Debugging Speech Recognition Systems. In: Text, Speech and Dialogue, Lecture Notes in Computer Science, 2010, Volume 6231/2010, 251-258, DOI: 10.1007/978-3-642-15760-8_32.
- M. Cernak
A Comparison of Decision Tree Classifiers for Automatic Diagnosis of Speech Recognition Errors. In Computing and Informatics. Vol. 29, no. 3 (2010), p. 489-501. ISSN 1335-9150.
- M. Cernak
Diagnostic Evaluation of Synthetic Speech Using Speech Recognition. In: The 16th International Congress on Sound and Vibration, Krakow, Poland 5-9 July 2009.
- M. Cernak
A Contribution of Intrinsic Speech Variabilities to Errors Done by Speech Recognition. In: Acoustics and Speech Processing, Bratislava 2008. ISBN 978-80-969202-8-0, pp. 149-156.
- M. Cernak and S. Darjaa
Noisy Speech Recognition Failure Diagnosis Using Minimum Message Length Decision Trees. In Proceedings of the 15th International Conference on Systems, Signals and Image Processing (IWSSIP), June 25-28, 2008, Bratislava, Slovak Republic
- M. Cernak, M. Benzeghiba and Ch. J. Wellekens
Diagnostics of Speech Recognition: on Evaluating Feature Set Performance. In: Proceedings of the 12th International Conference on Speech and Computer (SPECOM), vol. I, pp. 188-193. Moscow 2007. ISBN 6-7452-0110-X
- M. Cernak and Ch. J. Wellekens
Diagnostics of Speech Recognition Using Classification Phoneme Diagnostic Trees, CI 2006, special session on NLP, San Francisco, USA
- M. Cernak and Ch. J. Wellekens
Emotional aspects of intrinsic speech variabilities in automatic speech recognition, 11th SPECOM'2006, June 25-29, 2006, Saint-Petersburg, Russia
- M. Cernak
Unit selection speech synthesis in noise, ICASSP 2006, 31st IEEE ICASSP, May 14-19, 2006, Toulouse, France
- V. Tyagi, M. Benzheghiba, M. Cernak and Ch. Wellekens
Comparative Study of Different Features on OLLO Logatome Recognition Task, Speech Recognition and Intrinsic Variation Workshop, May 20, 2006 Toulouse, France
- M. Cernak and M. Trnka
Development of a Real-Time ASR System for Slovak Speechdat Database, 5th EURASIP Conference focused on Speech and Image Processing, Multimedia Communications and Services, June 2005, Smolenice, Slovakia
- T. Dutoit and M. Cernak
TTSBOX: A Matlab Toolbox for Teaching Text-to-Speech Synthesis, 30th IEEE ICASSP, March 2005, Philadelphia, PA, USA
- M. Cernak and M. Rusko
An evaluation of a synthetic speech using the PESQ measure, Proceedings of Forum Acusticum 2005, August 29 - September 2, Budapest, Hungary
- M. Cernak
The Use of Objective Quality Measures in Corpus-based Speech Synthesis, Ph.D Thesis, September 2004, Slovak University of Technology, Bratislava, Slovakia (in Slovak; an English abstract)
- M. Rusko, M. Trnka, S. Darjaa and M. Cernak
Slovak Speech Database for Experiments and Application Building in Unit Selection Speech Synthesis. In P. Sojka, I. Kopecek, and K. Pala, editors, Proceedings of TSD 2004. Springer, Brno, 2004
- M. Cernak
Slovko: A Phoneme-based Speech Re-sequencing System in MATLAB. In 3rd International Conference on Emerging Telecommunication Technologies and Applications, pages 95-98, 2004, Kosice, Slovakia
- M. Cernak and G. Rozinaj
Forward masking phenomenon in concatenative speech synthesis. In Proceedings of the 4th EURASIP Conference focused on Video/Image Processing and Multimedia Communications, July 2-5, 2003, pages 691-694 vol. 2, Zagreb, Croatia
- M. Cernak, M. Rusko, M. Trnka, and S. Darzagin
Data-driven Versus Knowledge-based Approaches to Orthoepic Transcription in Slovak. In International Conference on Emerging Telecommunication Technologies and Applications, pages 95-98, 2003, Kosice, Slovakia
- M. Cernak
Speech Synthesis in IVR Systems: A Proposal of the R&D Environment. In ELOSYS 2002, October 15-18, 2002, Trencín, Slovakia (in Slovak)
- M. Cernak
Speech Synthesis Research in Virtual Reality Systems. In Proceedings of the Int. Conference Research in Telecommunication Technology, September 17-19, 2002, Zilina, Slovakia
Technical Reports
- M. Cernak
DASR: A diagnostic tool for automatic speech recognition, Rapport de recherche RR-06-182, EURECOM, December 2006, Sophia-Antipolis, France
- M. Cernak
Command Speech Interface to Virtual Reality Applications, Technical Report, Virtual Reality Application Center, Iowa State University, June 2002, USA
Talks
- Failure Diagnosis Using Decision Trees: A Case Study of Human and Computer Speech Recognition
Telecommunications Forum, Telecommunication Research Center Vienna, September 2007, Austria
- Diagnostics of Speech Recognition Using Classification Phoneme Diagnostic Trees
DIVINES meeting, Avignon, June 2006, France
- Design, Implementation and Evaluation of the Robust Slovak Speech Synthesis in Noise
Talk at EURECOM, Sophia-Antipolis, October 2005, France
- Should Speech Synthesizers Have Ears?
Telecommunications Forum, Telecommunication Research Center Vienna, March 2003, Austria
Awards
- 2010 (team) Prize of the Sloval Academy of Sciences for the best applicable research project.
- A principal investigator of research grant assigned by Scientific Grant Agency of the Ministry of Education of Slovak Republic and the Academy of Sciences (2010 -- 2013) – Project: Automatic meeting speech recognition with application to courtroom recordings transcription.
- A principal investigator of research grant assigned by Scientific Grant Agency of the Ministry of Education of Slovak Republic and the Academy of Sciences (2007 -- 2010) – Project: Robust speech technologies for information systems and their diagnostics.
- Boeing sholarship awarded: visiting scientist at Virtual Reality Application Center, Iowa State University, 2002.
Software
People I have a great regard for
