Translation checking is a significant limiting factor on the pace of Bible translation efforts and the quality of the results. The intent of this project is to develop capacity and increase the pace while ensuring quality throughout the Bible translation community. Harnessing the latest artificial intelligence (AI) techniques, this project seeks to automatically and objectively assess multiple facets of translation quality and to validate the usability of those assessments with translation experts. In doing this, the project is working to create:
A set of shared tasks allows the AQuA community to coordinate our work, understand what contributions already exist, build on the work of others, and objectively compare results. After working with translation experts to gather and annotate data for our various tasks, we publish these as shared tasks on this site along with instructions for utilizing the data, submitting methods to be evaluated/validated, and a leaderboard of previously submitted methods. This site thus serves as the hub of our research and a starting point for developing assessment methods.
Comprehensibility 🤔 Similarity ⚖️ Naturalness 🗣️ Readability 📰 Alignment 📐 Automated Back Translation ⏮