Neural networks as classifiers in speech recognition systems
- 1 Dr.Sc. (Eng.), Professor, Head of the Department of Applied Mathematics and Mathematical Modeling, North-Caucasus Federal University, Stavropol, Russia
- 2,3 Post-graduate Student, Department of Applied Mathematics and Mathematical Modeling, North-Caucasus Federal University, Stavropol, Russia
Abstract
The paper analyzes the ability of using different neural networks for speech recognition systems. Neural networks offer massive parallelism to operate in real time and the ability to adapt, which creates a great potential for solving complex problems of speech recognition. Neural networks are applicable to all tasks necessary for speech recognition. However, for various subtasks requires different types of neural networks, which should be considered in establishing such systems.
How to cite
Червяков Н.И., Терновой Н.В., Гапочкин А.В. Использование нейронных сетей в качестве классификаторов в системах распознавания речи // Нейрокомпьютеры: разработка, применение. 2014. № 9. С. 20–24.
References
- Kohonen T., Makisara K., Saramaki T. Phono-topic Maps; Insightful Representation of Phonologic Features for Speech Recognition // Proc. 7th InL Conf. on Pattern Recognition, IEEE. New York. NY. 1984.
- Ryazanov V.V. Modeli i metody raspoznavaniya rechi. VTs RAN. 2011. 112 s.
- Vintsyuk T.K. Analiz, raspoznavanie i interpretatsiya rechevykh signalov. K.: Naukova dumka. 263 s.
- Chervyakov N.I., Lyakhov P.A. Realizatsiya KIKh-fil'trov v sisteme ostatochnykh klassov // Neyrokomp'yutery: razrabotka, primenenie. 2012. № 5. S. 16-25.