OpenNMT: Neural Machine Translation Toolkit

TL;DR

OpenNMT is an open-source toolkit designed for neural machine translation that emphasizes efficiency, modularity, and extensibility, supporting research and production use across multiple frameworks.

Contribution

It provides a flexible, well-documented platform for NMT research and deployment, integrating various model architectures and source modalities.

Findings

Used in multiple production MT systems

Modified for numerous research projects

Supports several deep learning frameworks

Abstract

OpenNMT is an open-source toolkit for neural machine translation (NMT). The system prioritizes efficiency, modularity, and extensibility with the goal of supporting NMT research into model architectures, feature representations, and source modalities, while maintaining competitive performance and reasonable training requirements. The toolkit consists of modeling and translation support, as well as detailed pedagogical documentation about the underlying techniques. OpenNMT has been used in several production MT systems, modified for numerous research papers, and is implemented across several deep learning frameworks.