Cairo: An Alignment Visualization Tool

Abstract

While developing a suite of tools for statistical machine translation research, we recognized the need for a visualization tool that would allow researchers to examine and evaluate specific word correspondences generated by a translation system. We developed Cairo to fill this need. Cairo is a free, open-source, portable, user-friendly, GUI-driven program written in Java that provides a visual representation of word correspondences between bilingual pairs of sentences, as well as relevant translation model parameters. This program can be easily adapted for visualization of correspondences in bi-texts based on probability distributions. 1. Background Cairo was originally developed with statistical transla-tion models of the sort developed at IBM in the early 1990s in mind (Berger et al., 1994). These models, referred to as the “Candide ” models, are based on the source-channel paradigm, where one language (the “source ” language) is interpreted as an encoded form of another (the “target ” lan-guage)1 (Weaver, 1949). A translation model is built us-ing iterative training on a sentence-aligned bi-text (Brown