Design of Russian Vocabulary Phonetic System Based on Cyclic Neural Network

DOI

10.2991/978-94-6463-264-4_62How to use a DOI?

Keywords

Circulatory neural network; Russian vocabulary; Phonetic system

Abstract

In order to improve the accuracy of Russian word pronunciation, this paper proposes a design of Russian word pronunciation system based on Recurrent neural network. Firstly, the traditional Russian phoneme set has been improved and designed to better express the pronunciation characteristics of words. On the basis of the new phoneme set, a Russian pronunciation dictionary containing 20000 words has been constructed. The Recurrent neural network (RNN) framework is used to implement this algorithm. This algorithm utilizes the encoder LSTM to convert Russian words into vector representations with fixed dimensions, and then converts the vectors into the target pronunciation sequence through the decoder LSTM. Through this method, the conversion process from Russian words to their corresponding pronunciations has been achieved. Finally, a fully functional Russian vocabulary pronunciation system was designed and implemented, including interactive word pronunciation and other functions. In the experiment, the system achieved significant results on the external word test set, with a word form accuracy rate of 75% and a phoneme accuracy rate of 95%. Compared to the traditional Phonetisaurus method, the performance is better. This means that the system has significant support and advantages in providing Russian pronunciation dictionary construction. This research has provided important progress for the development of Russian Speech processing, and a feasible solution for improving the accuracy and automation of Russian vocabulary pronunciation. In addition, the improved phoneme set design and the algorithm implementation based on Recurrent neural network also provide valuable reference and enlightenment for other languages' lexical phonetic systems.

Copyright

© 2024 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Yumin Xu
AU  - Liguo Liu
PY  - 2023
DA  - 2023/09/28
TI  - Design of Russian Vocabulary Phonetic System Based on Cyclic Neural Network
BT  - Proceedings of the 2023 3rd International Conference on Education, Information Management and Service Science (EIMSS 2023)
PB  - Atlantis Press
SP  - 542
EP  - 549
SN  - 2589-4900
UR  - https://doi.org/10.2991/978-94-6463-264-4_62
DO  - 10.2991/978-94-6463-264-4_62
ID  - Xu2023
ER  -