Email this article Contribution from the accuracy of phoneme recognition to the quality of automatic recognition of Russian speech
- Authors: Karpukhin I.A.1
-
Affiliations:
- Department of Computational Mathematics and Cybernetics
- Issue: Vol 40, No 2 (2016)
- Pages: 89-95
- Section: Article
- URL: https://journals.rcsi.science/0278-6419/article/view/176134
- DOI: https://doi.org/10.3103/S0278641916020060
- ID: 176134
Cite item
Open Access
Access granted
Subscription Access
Abstract
The dependence of the quality of automatic recognition of Russian speech on the accuracy of feature descriptions and their distribution model is considered. A list of eight phonemes is given whose accuracy of recognition makes the greatest contribution to the quality of speech recognition.
About the authors
I. A. Karpukhin
Department of Computational Mathematics and Cybernetics
Author for correspondence.
Email: karpuhini@yandex.ru
Russian Federation, Moscow, 119899
Supplementary files
