Petr Pollák - ČVUT FEL K331 |
|
Borský, M. - Mizera, P. - Pollák, P.: Noise and Channel Normalized Cepstral Features for Far-Speech Recognition. In Speech and Computer. Cham: Springer International Publishing AG, 2013, p. 241-248. ISSN 0302-9743.ISBN 978-3-319-01930-7.
Borský, M. - Pollák, P.: Optimized State-Tying for Triphone-Based HMMs under Training Data Deficiency. In Applied Electronics - 2013 International Conference on Applied Electronics. Pilsen: University of West Bohemia, 2013, p. 45-48. ISSN 1803-7232.ISBN 978-80-261-0166-6.
Borský, M. - Pollák, P.: Various Approaches of Small Vocabulary Speech Recognizer Implementation Using HTK Toolkit. In POSTER 2013 - 17th International Student Conference on Electrical Engineering. Prague: Czech Technical University, 2013, p. 1-5. ISBN 978-80-01-05242-6.
Mizera, P. - Pollák, P.: Accuracy of HMM-Based Phonetic Segmentation Using Monophone or Triphone Acoustic Model. In Applied Electronics - 2013 International Conference on Applied Electronics. Pilsen: University of West Bohemia, 2013, p. 181-184. ISSN 1803-7232.ISBN 978-80-261-0166-6.
Rajnoha, J. - Pollák, P. (supervisor): Modelování neřečových událostí pro rozpoznávání řeči v reálných podmínkách. [Doktorská práce (Ph.D.)]. 2013. 133 s.