This dataset is a self-authored synthetic benchmark for multilingual video localization workflows. It focuses on the timing pressure that appears when subtitle segments are translated across languages and then reviewed for dubbing fit, subtitle-window preservation, and lip-sync risk. The package is designed for repository-safe experimentation and documentation. It does not contain third-party video, third-party audio… See the full description on the dataset page: https://huggingface.co/datasets/hellohihiloy789/ai-translate-video.
