The first neural machine translation system for the Erzya language

TL;DR

This paper introduces the first neural machine translation system for Erzya, an endangered language, including datasets, models, and tools, achieving promising translation quality for Erzya-Russian and initial multilingual adaptation.

Contribution

It presents the first neural translation system for Erzya, along with datasets, models, and language tools, enabling future research and development for this low-resource language.

Findings

BLEU scores of 17 and 19 for Erzya-Russian translation

Over half of translations rated acceptable by native speakers

Multilingual adaptation shows low quality without additional data

Abstract

We present the first neural machine translation system for translation between the endangered Erzya language and Russian and the dataset collected by us to train and evaluate it. The BLEU scores are 17 and 19 for translation to Erzya and Russian respectively, and more than half of the translations are rated as acceptable by native speakers. We also adapt our model to translate between Erzya and 10 other languages, but without additional parallel data, the quality on these directions remains low. We release the translation models along with the collected text corpus, a new language identification model, and a multilingual sentence encoder adapted for the Erzya language. These resources will be available at https://github.com/slone-nlp/myv-nmt.