Opus Mt En Gil
opus-mt-en-gil is a machine translation model based on the Transformer architecture, specifically designed for translating English (en) to Gilbertese (gil).
Downloads 30
Release Time : 3/2/2022
Model Overview
This model is a machine translation model trained on the OPUS dataset, utilizing the Transformer-align architecture, supporting English to Gilbertese translation tasks.
Model Features
Based on OPUS Dataset
The model is trained on the OPUS dataset, ensuring broad coverage for translation quality.
Transformer-align Architecture
Utilizes the Transformer-align architecture, optimizing translation alignment effects.
Standardized Preprocessing
Input text undergoes standardization and SentencePiece preprocessing, enhancing model robustness.
Model Capabilities
English to Gilbertese Text Translation
Use Cases
Language Translation
English to Gilbertese Document Translation
Translate English documents into Gilbertese, suitable for multilingual document processing.
Achieved scores of BLEU 38.8 and chr-F 0.604 on the JW300.en.gil test set.
Featured Recommended AI Models
Š 2025AIbase