Dubbing in Practice: A Large Scale Study of Human Localization With Insights for Automatic Dubbing
This research addresses the challenge of improving automatic dubbing systems for video localization by providing data-driven insights that challenge existing assumptions in both human and machine-learning literature.
The study tackled the problem of understanding human dubbing practices by analyzing a large corpus of 319.57 hours of professionally produced video, finding that vocal naturalness and translation quality are more critical than isometric and lip-sync constraints, and highlighting the influence of source-side audio on speech characteristics and semantic transfer.
We investigate how humans perform the task of dubbing video content from one language into another, leveraging a novel corpus of 319.57 hours of video from 54 professionally produced titles. This is the first such large-scale study we are aware of. The results challenge a number of assumptions commonly made in both qualitative literature on human dubbing and machine-learning literature on automatic dubbing, arguing for the importance of vocal naturalness and translation quality over commonly emphasized isometric (character length) and lip-sync constraints, and for a more qualified view of the importance of isochronic (timing) constraints. We also find substantial influence of the source-side audio on human dubs through channels other than the words of the translation, pointing to the need for research on ways to preserve speech characteristics, as well as semantic transfer such as emphasis/emotion, in automatic dubbing systems.