Journal of Siberian Federal University. Engineering & Technologies / On Applicability of Recurrent Neural Networks to Language Modelling for Inflective Languages

Full text (.pdf)
Issue
Journal of Siberian Federal University. Engineering & Technologies. 2016 9 (8)
Authors
Kudinov, Mikhail S.
Contact information
Kudinov, Mikhail S.: Samsung R&D Center Russia 1 corp., 12, Dvintsev, Moscow, 127018, Russia;
Keywords
language models; recurrent neural network; inflected languages; speech recognition
Abstract

Standard version of recurrent neural network language model (RNNLM) has shown modest results in language modelling of Russian. In this paper we present a special modification of RNNLM making separate predictions of lemmas and morphology. New model shows superior results compared to Knesser-Ney language model both in perplexity and in ranking experiment. At the same time morphology integration has not shown any improvement

Pages
1291-1301
Paper at repository of SibFU
https://elib.sfu-kras.ru/handle/2311/30351

Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).