LoginRegistration
For instance: The Scientific Opinion
About consortium subscription Contacts
(812) 4095364 Non-commercial partnership
St. Petersburg
university
consortium

Articles

"Humanities and Science University Journal" № 7, 2014

Use of Articulatory Data in Tasks Related to the Automated Russian Speech Recognition

D. A. Kocharov
Price: 50 руб.
The article represents the results of tests in the automated interpretation of sounds in Russian speech by using articulatory data as differential indicators. The system utilizes a triple tier of recurrent neural network as a classifi er. Test results revealed a
high effi cacy of using articulatory data in resolving the task of automated identifi cation. Overall effi ciency of this system reaches 69%, while about 33% of the sounds are identifi ed with 100% accuracy.
Keywords: electromagnetic articulography, automatic speech recognition, speech processing, experimental phonetics, linguistics.
REFERENCES
1. Kocharov D. A., Glotova O. N. Obzor tekhnologii elektromagnitnoy artikulografi i
dlya issledovaniya i modelirovaniya protsessov porozhdeniya rechi // Nauchnoe mnenie. SPb., 2012. № 12. S. 47–53.
2. Kocharov D. A., Glotova O. N. Korpus artikulyatornykh dannykh russkoy
rechi // Trudy mezhdunarodnoy konferentsii «Korpusnaya lingvistika — 2013». SPb.:
S.-Peterburgskiy gos. universitet, 2013. S. 328–335.
3. Graves A., Mohamed A. and Hinton G. E. Speech Recognition with Deep Recurrent
Neural Networks // In IEEE International Conference on Acoustic Speech and
Signal Processing (ICASSP 2013) Vancouver, 2013.
4. Kaburagi T., Wakamiya K., Honda M. Three-dimensional electromagnetic articulography: A measurement principle. The Journal of the Acoustical Society of America. Vol. 118 (2005). Р. 428–443. 2005.
5. King S., Frankel J., Livescu K., McDermott E., Richmond K. and Wester M. Speech
production knowledge in automatic speech recognition. Journal of the Acoustical Society of America, 121(2): 723–742, February 2007.
6. Mohamed A., Dahl G. E. and Hinton G. E. Deep belief networks for phone recognition
// Proc. NIPS Workshop Deep Learning Speech Recognition Relative Applications,
2009.
7. Moon T., Stirling W. Mathematical Methods and Algorithms for Signal Processing,
Prentice Hall, 1999.
8. Richmond K., Ling Z., Yamagishi J., and Uria B. On the evaluation of inversion
mapping performance in the acoustic domain // Proc. Interspeech, pp. 1012–1016,
Lyon, 2013.
9. Robinson A. J. An Application of Recurrent Nets to Phone Probability Estimation
// IEEE Transactions on Neural Networks. Vol. 5, no. 2/ P. 298–305, 1994.
Price: 50 рублей
To order