Polish LVCSR in the Janus system. Preliminary results for the SpeeCon database

• Autorzy: Marasek K.
• Języki publikacji: EN

Abstrakty

EN

This paper describes the development of the LVCSR (Large Vocabulary Continuous Speech Recognition) system for Polish, using the Janus system developed at the University Karlsruhe/Carnegie Mellon University. The system has been tested on the selected material from the SpeeCon database. Test results for sentences read by 16 speakers are given. The system shows good performance and can be used as a basis for further development of modern speech recognition technology for Polish.

Słowa kluczowe

EN

speech recognition; LVCSR; Janus JRTk

• Wydawca: Instytut Podstawowych Problemów Techniki PAN ; Komitet Akustyki PAN ; Polskie Towarzystwo Akustyczne
• Czasopismo: Archives of Acoustics
• Rocznik: 2007
• Tom: Vol. 32, No. 1
• Strony: 119--126
• Opis fizyczny: Bibliogr. 12 poz., tab.

Twórcy

autor

Marasek K.

• Polish-Japanese Institute of Information Technology, Koszykowa 86, 02-008 Warszawa, Poland,

Bibliografia

• [1] DE MORI, Spoken dialogues with computers: Signal processing and its applications, Academic Press, Hardcover, 1998, ISBN 0122090551.
• [2] FINKE M., GEUTNER P., HILD H., KEMP T., RIES K., WESTPHAL M., The Karlsruhe. VERBMOBIL speech recognition engine, Proceedings of the IEEE ASSP Conf., vol. 1, pp. 83.86 Munich 1997.
• [3] HUANG, ACERO, HON, Spoken language processing, Prentice Hall, 2001, ISBN 0130226165.
• [4] MATSOUKAS S., SCHWARTZ R., JIN H., NGUYEN L., Practical implementations of speakeradaptive training, proceedings of the 1997 DARPA speech recognition workshop, Chantilly, Virginia, USA, February 2.5, 1997.
• [5] SOLTAU H., METZE F., FÜGEN CH., WAIBEL A., A one pass-decoder based on polymorphic linguistic context assignment, Proceedings of the ASRU Workshop, Madonna di Campiglio, Italy, pp. 214.217, 2001.
• [6] FRITSCH J., ROGINA I., The bucket box intersection (BBI) algorithm for fast approximative evaluation of diagonal mixture Gaussians, Proceedings of ICASSP 96, Atlanta, Vol. 2, pp. 837.840, 1996.
• [7] ROACH P. et al., Babel: An eastern european multi-language database, proceedings of ICSLP-96, Philadelphia, pp. 1982.1986, 1996.
• [8] MARASEK K., Large vocabulary continuous speech recognition system for Polish, Archives of Acoustics, 28, 4, 293.303 (2003).
• [9] MARASEK K., GUBRYNOWICZ R., Mutlilevel annotation in SpeeCon Polish speech database, IMTCI (International Workshop on Intelligent Media Technology for Communicative Inteligence), Warszawa, Lecture Notes in Computer Science, Springer, pp. 58.67, 2004.
• [10] IslSystem Documentation, Example Training/Testing Setup for Use with JRTk, The Ibis-gang, 22. November 2002, v0.1, Interactive Systems Labs, University of Karlsruhe, Germany, 2002.
• [11] STOLCKE A., SRILM . An extensible language modeling toolkit, Proc. of ICSLP 2002, Denver, Colorado, pp. 901.904, 2002.
• [12] WOODLAND P. C., ODELL J. J., VALTCHEV V., YOUNG S. J., Large vocabulary continuous speech recognition using HTK, Proceedings ICASSP'94, Adelaide, pp. 125.128, 1994.