Rethinking dubbing workflows: The tentative role of pre-editing in machine-translated content
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
The DubTA project investigates the integration of neural machine translation into dubbing workflows. Given that audiovisual translation has historically been less permeable to machine translation due to its inherently creative and multimodal nature, this article first reviews the recent increase in automation technologies within audiovisual translation. This context motivated the DubTA project, whose methodology and preliminary findings are presented here. The primary goal of DubTA is to evaluate the feasibility of incorporating machine translation into dubbing processes. To this end, raw output generated by two machine translation engines was analyzed, with errors systematically categorized using a custom taxonomy to identify areas suitable for potential pre-editing in fictional dubbing scripts. Based on these findings, the project explores the possibility of developing pre-editing guidelines that could help enhance machine translation output and facilitate dubbing workflows by reducing the need for extensive post-editing. While promising, these results highlight the need for further research to refine these preliminary guidelines and assess their impact on diverse dubbing scenarios.
Fields of Study & Subjects
Translation Studies and Practices
Subtitles and Audiovisual Media
Natural Language Processing Techniques
Publication details
- Journal / source
- Cadernos de Tradução
- Volume
- 46
- Pages
- 1–25
- Publication date
- 26 March 2026
- Language
- EN