Opus-mt-vi-en Open-source Machine Translation Model - Free Translation from Vietnamese (including Chinese character variants) to English

Opus Mt Vi En

Developed by Helsinki-NLP

A Transformer-based machine translation model from Vietnamese to English, developed by Helsinki-NLP, supporting standardized Vietnamese (including Han character variants) to English translation tasks.

Machine Translation

Transformers

Supports Multiple LanguagesOpen Source License:Apache-2.0 #Transformer architecture #Vietnamese to English translation #SentencePiece tokenization

Downloads 34.82k

Release Time : 3/2/2022

Model Overview

This model is specifically designed for Vietnamese to English translation, utilizing SentencePiece tokenization for preprocessing, suitable for general text translation scenarios.

Model Features

Standardized preprocessing

Uses standardization and SentencePiece tokenization (spm32k) to process input text, ensuring translation quality.

High-performance translation

Achieves high scores of 42.8 BLEU and 0.608 chr-F on the Tatoeba-test.vie.eng dataset.

Support for Han character variants

Capable of handling Han character variants in Vietnamese, accommodating diverse input texts.

Model Capabilities

Vietnamese to English text translation

Supports standardized text input

Handles Han character variants

Use Cases

Text translation

Daily conversation translation

Translates Vietnamese daily conversations into English for cross-language communication.

Achieves a high score of 42.8 BLEU on the test dataset.

Document translation

Translates Vietnamese document content into English for business or academic purposes.

Supports standardized text input to ensure translation quality.

🚀 Vie-Eng Translation Model

This is a Vietnamese-to-English translation model. It provides high - quality translation services and has been tested on specific benchmarks.

🚀 Quick Start

This Vie - Eng translation model is designed to translate Vietnamese text into English. You can download the original weights and test set translations from the provided links to start using and evaluating the model.

✨ Features

Language Pair: Translates from Vietnamese (vie) to English (eng).
Model Type: transformer - align
Pre - processing: Normalization + SentencePiece (spm32k, spm32k)

📦 Installation

To use this model, you need to download the original weights: opus - 2020 - 06 - 17.zip

💻 Usage Examples

You can use the downloaded weights to perform translations. Although no specific code examples are provided here, you can use relevant machine - learning frameworks to load the model and conduct translation tasks.

📚 Documentation

General Information

Source Group: Vietnamese
Target Group: English
OPUS Readme: vie - eng

Benchmarks

Testset	BLEU	chr - F
Tatoeba - test.vie.eng	42.8	0.608

System Info

Property	Details
hf_name	vie - eng
Source Languages	vie
Target Languages	eng
OPUS Readme URL	vie - eng
Original Repo	Tatoeba - Challenge
Tags	['translation']
Languages	['vi', 'en']
Source Constituents	{'vie', 'vie_Hani'}
Target Constituents	{'eng'}
Source Multilingual	False
Target Multilingual	False
Pre - processing	normalization + SentencePiece (spm32k, spm32k)
URL Model	opus - 2020 - 06 - 17.zip
URL Test Set	opus - 2020 - 06 - 17.test.txt
Source Alpha3	vie
Target Alpha3	eng
Short Pair	vi - en
chrF2 Score	0.608
BLEU	42.8
Brevity Penalty	0.955
Ref Len	20241.0
Source Name	Vietnamese
Target Name	English
Train Date	2020 - 06 - 17
Source Alpha2	vi
Target Alpha2	en
Prefer Old	False
Long Pair	vie - eng
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