Between Flexibility and Consistency: Joint Generation of Captions and Subtitles
This work addresses the need for consistent and high-quality subtitles in multilingual scenarios, though it is incremental in improving existing speech translation methods.
The paper tackled the joint generation of source captions and target subtitles in speech translation, focusing on consistency in structure and lexical content, and introduced new metrics for evaluation. The results showed that joint decoding improved performance and consistency while maintaining flexibility for language-specific norms.
Speech translation (ST) has lately received growing interest for the generation of subtitles without the need for an intermediate source language transcription and timing (i.e. captions). However, the joint generation of source captions and target subtitles does not only bring potential output quality advantages when the two decoding processes inform each other, but it is also often required in multilingual scenarios. In this work, we focus on ST models which generate consistent captions-subtitles in terms of structure and lexical content. We further introduce new metrics for evaluating subtitling consistency. Our findings show that joint decoding leads to increased performance and consistency between the generated captions and subtitles while still allowing for sufficient flexibility to produce subtitles conforming to language-specific needs and norms.