Enhancing Translation Efficiency with AI: Predicting TER Scores and Optimizing TM Leverage

Track: Advanced Localization Management | AL9 |
Friday, June 7, 2024, 3:45pm – 4:30pm
Held in: Herbert room
Rimi Endo - Centific 
Eric Woelfel - Centific

In the fast-paced world of localization, ensuring both efficiency and accuracy in translation is paramount. Traditional methods of providing translation quotes and ensuring high translation memory (TM) leverage often fall short, especially in the absence of comprehensive TM databases. In this session, the presenters explore the innovative use of Artificial Intelligence (AI) to predict translation edit rate (TER) scores for generating more precise translation quotes. Additionally, they delve into how AI can be utilized to refine high TM leverage cases, automatically correcting punctuation, typos, and grammar errors to potentially upgrade them to a proposed 100% match. By integrating AI into these facets of the localization workflow, companies can significantly enhance the precision of their translation quotes and boost the efficacy of their TM leverage, leading to improved productivity, cost savings, and ultimately, higher-quality translations.


  • Participants will gain insights into the practical applications of AI in refining translation workflows, specifically in predicting TER scores and optimizing TM leverage.
  • The session will provide concrete examples and case studies demonstrating the effectiveness of these AI-driven approaches, highlighting the potential time and cost savings, as well as the improvements in translation accuracy.
  • Attendees will leave with a comprehensive understanding of how these technologies can be seamlessly integrated into their existing localization processes, fostering a more efficient, accurate, and cost-effective translation environment.