Moses is a statistical machine translation system for building translation models from bilingual and monolingual text. It is closely associated with phrase-based machine translation, in which correspondences between sequences of words are learned from aligned parallel corpora.
The software distinguishes model preparation from translation. A training pipeline processes aligned examples to estimate phrase and language-model information; the decoder searches among possible translations to produce likely target-language sentences. Researchers can inspect the parts of the pipeline, alter model settings and compare different approaches.
Moses remains useful when reproducing research or maintaining statistical machine translation systems. Neural machine translation now dominates many production applications, so the project should be described as an established statistical framework rather than as the prevailing technique used by major online translation services.
This is free and open source software.
Key Features
- Trains phrase-based translation models using aligned bilingual text.
- Includes hierarchical phrase-based and syntax-based decoding methods.
- Supports factored translation models incorporating word-level annotations.
- Provides a decoder that searches for probable translations under a trained model.
- Can process word lattices and confusion networks as decoder input.
- Includes feature functions for scoring alternative translation hypotheses.
- Supports tuning and evaluation workflows for translation experiments.
- Offers scripts and utilities for working with parallel and monolingual corpora.
- Includes an Experiment Management System for organizing runs.
- Exposes configurable decoding behaviour for research comparisons.
- Supports phrase tables, reordering models and feature weight combinations.
- Can combine a translation model with an external statistical language model.
- Offers multiple decoding hypotheses for inspecting uncertain translations.
- Includes regression tests for decoder behaviour and model experiments.
Website: github.com/moses-smt/mosesdecoder
Support:
Developer: Hieu Hoang, Philipp Koehn and contributors
License: GNU Lesser General Public License v2.1
Moses is written in Perl and C++. Learn Perl with our recommended free books and free tutorials. Learn C++ with our recommended free books and free tutorials.
Related Software
| Natural Language Processing | |
|---|---|
| PyTorch-Transformers | Library of state-of-the-art pre-trained models |
| Natural Language Toolkit | Suite of open source Python modules, data sets and tutorials |
| Stanford CoreNLP | Extensible annotation-based NLP pipeline |
| spaCy | Industrial strength natural language processing |
| scikit-learn | Machine learning library for Python |
| Gensim | Python-based vector space modeling and topic modeling toolkit |
| flair | Simple framework for state-of-the-art NLP |
| Apache OpenNLP | Machine learning based toolkit |
| DL4J | Deploy and train deep learning models |
| Apache Lucene | Full-featured information retrieval software library |
| UIMA | Implementation of the UIMA specification |
| tidytext | Text mining using dplyr, ggplot2, and other tidy tools |
| text2vec | Framework with API for text analysis and NLP |
| quanteda | R package for Quantitative Analysis of Textual Data |
| Moses | Statistical machine translation system |
Read our verdict in the software roundup.
| C++ Natural Language Processing Tools | |
|---|---|
| text2vec | Framework with API for text analysis and natural language processing |
| Moses | Statistical machine translation system |
| TiMBL | Tilburg Memory-Based Learner |
| MITIE | MIT Information Extraction |
| MeTA | Modern C++ data sciences toolkit |
| Colibri Core | Efficient n-gram & skipgram modelling on text corpora |
| CRF++ | Yet Another CRF toolkit |
| BLLIP Parser | Statistical natural language parser |
Read our verdict in the software roundup.
 LinuxLinks helps you discover the best free and open source software for Linux. Explore our extensive collection of recommended applications, including alternatives to popular proprietary software.We also publish software reviews, Linux tutorials, hardware reviews, free programming books and articles covering the wider open source community. Browse our Linux resources to discover more. Know a useful free and open source Linux application we haven't covered? Suggest a program for inclusion. Has your software been featured on LinuxLinks? Display one of our Featured on LinuxLinks badges on your website or GitHub README. |
Please read our Comment Policy before commenting.