Opus-mt-fr-ar Open-source Translation Model - Achieve Accurate Translation from French to Multiple Arabic Dialects for Free

Opus Mt Fr Ar

Developed by Helsinki-NLP

This is a Transformer-based machine translation model for French to Arabic, supporting multiple Arabic dialect variants.

Machine Translation

Transformers

Supports Multiple LanguagesOpen Source License:Apache-2.0 #French-Arabic multidialectal translation #SentencePiece tokenization #Low-resource language support

Downloads 2,832

Release Time : 3/2/2022

Model Overview

Developed by the Helsinki-NLP team, this model is specifically designed for French to Arabic text translation tasks, supporting Modern Standard Arabic and various dialect variants.

Model Features

Multidialect support

Supports Modern Standard Arabic and various dialect variants translation

Preprocessing optimization

Utilizes normalization and SentencePiece tokenization for text preprocessing

Language identifiers

Implements multidialect distinction by adding target language identifiers at the beginning of sentences

Model Capabilities

French to Arabic text translation

Multidialect Arabic translation

Cross-language text conversion

Use Cases

Language translation

French content localization

Translate French content into Arabic for Middle Eastern regions

BLEU 14.4/chr-F 0.439

Multidialect translation

Supports translation needs for various Arabic dialects

🚀 French - Arabic Translation Model

This is a French to Arabic translation model, providing a solution for language translation between French and Arabic.

✨ Features

Language Support: Supports translation from French (fra) to multiple Arabic variants including apc, ara, arq, arq_Latn, ary, arz.
Model Type: Utilizes the transformer model architecture.
Pre - processing: Applies normalization and SentencePiece (spm32k,spm32k) for pre - processing.
Initial Token Requirement: Requires a sentence initial language token in the form of >>id<< (where id is a valid target language ID).

📦 Installation

The installation mainly involves downloading the model weights. You can download the original weights from opus - 2020 - 07 - 03.zip.

📚 Documentation

Model Details

Property	Details
Model Type	transformer
Source Language	fra (French)
Target Languages	apc, ara, arq, arq_Latn, ary, arz
Pre - processing	normalization + SentencePiece (spm32k,spm32k)
Download Link (Original Weights)	opus - 2020 - 07 - 03.zip
Test Set Translations	opus - 2020 - 07 - 03.test.txt
Test Set Scores	opus - 2020 - 07 - 03.eval.txt

Benchmarks

testset	BLEU	chr - F
Tatoeba - test.fra.ara	14.4	0.439

System Info

hf_name: fra - ara
source_languages: fra
target_languages: ara
opus_readme_url: https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/fra-ara/README.md
original_repo: Tatoeba - Challenge
tags: ['translation']
languages: ['fr', 'ar']
src_constituents: {'fra'}
tgt_constituents: {'apc', 'ara', 'arq_Latn', 'arq', 'afb', 'ara_Latn', 'apc_Latn', 'arz'}
src_multilingual: False
tgt_multilingual: False
prepro: normalization + SentencePiece (spm32k,spm32k)
url_model: https://object.pouta.csc.fi/Tatoeba-MT-models/fra-ara/opus-2020-07-03.zip
url_test_set: https://object.pouta.csc.fi/Tatoeba-MT-models/fra-ara/opus-2020-07-03.test.txt
src_alpha3: fra
tgt_alpha3: ara
short_pair: fr - ar
chrF2_score: 0.439
bleu: 14.4
brevity_penalty: 1.0
ref_len: 7956.0
src_name: French
tgt_name: Arabic
train_date: 2020 - 07 - 03
src_alpha2: fr
tgt_alpha2: ar
prefer_old: False
long_pair: fra - ara
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