Opus-mt-ko-ru Open Source Machine Translation Model - Free and Efficient Translation from Korean to Russian

Opus Mt Ko Ru

Developed by Helsinki-NLP

This is a Korean to Russian machine translation model based on the transformer-align architecture, released by the Tatoeba-Challenge project.

Machine Translation

Transformers

Supports Multiple LanguagesOpen Source License:Apache-2.0 #Korean-Russian Translation #Multilingual Support #SentencePiece Preprocessing

Downloads 175

Release Time : 3/2/2022

Model Overview

This model is specifically designed for Korean (including Hangul and romanized text) to Russian translation tasks, utilizing standardized preprocessing and SentencePiece tokenization techniques.

Model Features

Multi-format Korean Support

Supports translation of Hangul, romanized text, and mixed Hanja inputs.

Standardized Preprocessing

Employs standardization and SentencePiece tokenization to enhance translation quality.

Alignment Mechanism

Uses the transformer-align architecture to achieve word alignment between source and target languages.

Model Capabilities

Korean to Russian text translation

Multi-format Korean text processing

Use Cases

Cross-language Communication

Korean-Russian Chat Translation

Used for instant messaging translation between Korean and Russian users.

BLEU 30.3, chr-F 0.514

Content Localization

Korean Content Russianization

Translates Korean news, blogs, and other content into Russian.

🚀 kor-rus Translation Model

This project focuses on the translation between Korean and Russian, providing a high - quality translation model and related evaluation data.

✨ Features

Language Pair: It supports translation from Korean to Russian.
Model Type: The transformer - align model is used for translation.
Pre - processing: The pre - processing includes normalization and SentencePiece (spm32k, spm32k).

📦 Installation

No specific installation steps are provided in the original document.

📚 Documentation

Model Information

Source Group: Korean
Target Group: Russian
OPUS Readme: kor - rus
Model: transformer - align
Source Language(s): kor_Hang, kor_Latn
Target Language(s): rus
Pre - processing: normalization + SentencePiece (spm32k, spm32k)
Download Original Weights: opus - 2020 - 06 - 17.zip
Test Set Translations: opus - 2020 - 06 - 17.test.txt
Test Set Scores: opus - 2020 - 06 - 17.eval.txt

Benchmarks

Testset	BLEU	chr - F
Tatoeba - test.kor.rus	30.3	0.514

System Info

Property	Details
hf_name	kor - rus
source_languages	kor
target_languages	rus
opus_readme_url	https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/kor-rus/README.md
original_repo	Tatoeba - Challenge
tags	['translation']
languages	['ko', 'ru']
src_constituents	{'kor_Hani', 'kor_Hang', 'kor_Latn', 'kor'}
tgt_constituents	{'rus'}
src_multilingual	False
tgt_multilingual	False
prepro	normalization + SentencePiece (spm32k, spm32k)
url_model	https://object.pouta.csc.fi/Tatoeba-MT-models/kor-rus/opus-2020-06-17.zip
url_test_set	https://object.pouta.csc.fi/Tatoeba-MT-models/kor-rus/opus-2020-06-17.test.txt
src_alpha3	kor
tgt_alpha3	rus
short_pair	ko - ru
chrF2_score	0.514
bleu	30.3
brevity_penalty	0.961
ref_len	1382.0
src_name	Korean
tgt_name	Russian
train_date	2020 - 06 - 17
src_alpha2	ko
tgt_alpha2	ru
prefer_old	False
long_pair	kor - rus
helsinki_git_sha	480fcbe0ee1bf4774bcbe6226ad9f58e63f6c535
transformers_git_sha	2207e5d8cb224e954a7cba69fa4ac2309e9ff30b
port_machine	brutasse
port_time	2020 - 08 - 21 - 14:41

📄 License

This project is licensed under the Apache - 2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご