Opus-mt-zh-de Open-source Translation Model - Supports Heavyweight Chinese-German Translation of Multiple Chinese Dialect Variants

Opus Mt Zh De

Developed by Helsinki-NLP

This is a machine translation model from Chinese to German based on the transformer-align architecture, supporting translation from various Chinese dialect variants such as Mandarin and Cantonese to German.

Machine Translation

Transformers

Supports Multiple LanguagesOpen Source License:Apache-2.0 #Chinese-German Translation #Multi-dialect Support #SentencePiece Tokenization

Downloads 472

Release Time : 3/2/2022

Model Overview

This model is specifically designed for Chinese-to-German text translation tasks, employing standardized preprocessing and SentencePiece tokenization techniques, suitable for various Chinese dialect variants.

Model Features

Multi-dialect Support

Supports translation from various Chinese dialect variants such as Mandarin, Cantonese, and Wu Chinese.

Standardized Preprocessing

Employs standardized preprocessing and SentencePiece tokenization techniques to enhance translation quality.

High Performance

Achieves 32.1 BLEU score and 0.522 chr-F score on the Tatoeba test set.

Model Capabilities

Chinese-to-German Text Translation

Multi-dialect Chinese Processing

Standardized Text Preprocessing

Use Cases

Language Services

Document Translation

Translate Chinese documents into German

Achieves 32.1 BLEU score on the Tatoeba test set

Multi-dialect Translation

Supports translation from dialects like Cantonese to German

Education

Language Learning Assistance

Helps Chinese learners understand German texts

🚀 Chinese - German Translation Model

This is a Chinese - German translation model that provides high - quality translation services between the two languages. It uses advanced techniques and has been tested and evaluated with specific benchmarks.

✨ Features

Supports translation from multiple Chinese variants (e.g., Mandarin, Cantonese, etc.) to German.
Utilizes the transformer - align model for translation.
Applies pre - processing techniques of normalization and SentencePiece.

📦 Installation

The original weights of the model can be downloaded from opus - 2020 - 06 - 17.zip.

📚 Documentation

zho - deu

Source Group: Chinese
Target Group: German
OPUS Readme: zho - deu
Model: transformer - align
Source Language(s): cmn, cmn_Bopo, cmn_Hang, cmn_Hani, cmn_Hira, cmn_Kana, cmn_Latn, lzh_Hani, wuu_Hani, yue_Hani
Target Language(s): deu
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.zho.deu	32.1	0.522

System Info

Property	Details
hf_name	zho - deu
source_languages	zho
target_languages	deu
opus_readme_url	https://github.com/Helsinki-NLP/Tatoeba-Challenge/tree/master/models/zho-deu/README.md
original_repo	Tatoeba - Challenge
tags	['translation']
languages	['zh', 'de']
src_constituents	{'cmn_Hans', 'nan', 'nan_Hani', 'gan', 'yue', 'cmn_Kana', 'yue_Hani', 'wuu_Bopo', 'cmn_Latn', 'yue_Hira', 'cmn_Hani', 'cjy_Hans', 'cmn', 'lzh_Hang', 'lzh_Hira', 'cmn_Hant', 'lzh_Bopo', 'zho', 'zho_Hans', 'zho_Hant', 'lzh_Hani', 'yue_Hang', 'wuu', 'yue_Kana', 'wuu_Latn', 'yue_Bopo', 'cjy_Hant', 'yue_Hans', 'lzh', 'cmn_Hira', 'lzh_Yiii', 'lzh_Hans', 'cmn_Bopo', 'cmn_Hang', 'hak_Hani', 'cmn_Yiii', 'yue_Hant', 'lzh_Kana', 'wuu_Hani'}
tgt_constituents	{'deu'}
src_multilingual	False
tgt_multilingual	False
prepro	normalization + SentencePiece (spm32k, spm32k)
url_model	https://object.pouta.csc.fi/Tatoeba-MT-models/zho-deu/opus-2020-06-17.zip
url_test_set	https://object.pouta.csc.fi/Tatoeba-MT-models/zho-deu/opus-2020-06-17.test.txt
src_alpha3	zho
tgt_alpha3	deu
short_pair	zh - de
chrF2_score	0.522
bleu	32.1
brevity_penalty	0.9540000000000001
ref_len	19102.0
src_name	Chinese
tgt_name	German
train_date	2020 - 06 - 17
src_alpha2	zh
tgt_alpha2	de
prefer_old	False
long_pair	zho - deu
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