Information-theoretic analysis of efficiency of the phonetic encoding–decoding method in automatic speech recognition
- Authors: Savchenko V.V.1, Savchenko A.V.2
-
Affiliations:
- Nizhny Novgorod State Linguistic University
- National Research University Higher School of Economics
- Issue: Vol 61, No 4 (2016)
- Pages: 430-435
- Section: Theory and Methods of Signal Processing
- URL: https://journals.rcsi.science/1064-2269/article/view/196922
- DOI: https://doi.org/10.1134/S1064226916040112
- ID: 196922
Cite item
Abstract
A words phonetic decoding method in automatic speech recognition is considered. The properties of Kullback–Leibler divergence are used to synthesize the estimation of the distribution of divergence between minimum speech units (e.g., single phonemes) inside a single class. It is demonstrated that the minimum variance of the intraphonemic divergence is reached when the phonetic database is tuned to the voice of a single speaker. The estimations are proven by experimental results on the recognition of vowel sounds and isolated words of Russian language.
About the authors
V. V. Savchenko
Nizhny Novgorod State Linguistic University
Email: avsavchenko@hse.ru
Russian Federation, ul. Minina 31a, Nizhny Novgorod, 603155
A. V. Savchenko
National Research University Higher School of Economics
Author for correspondence.
Email: avsavchenko@hse.ru
Russian Federation, Bol’shaya Pecherskaya ul. 25/12, Nizhny Novgorod, 603155