Between Flexibility and Consistency: Joint Generation of Captions and Subtitles

TL;DR

This paper explores joint generation of captions and subtitles in speech translation, demonstrating that combined decoding improves output quality and consistency while maintaining flexibility for language-specific norms.

Contribution

It introduces a joint decoding approach for captions and subtitles in speech translation and proposes new metrics for evaluating their consistency.

Findings

Joint decoding enhances caption and subtitle consistency.

The approach improves translation quality and adherence to language norms.

New metrics effectively measure subtitling consistency.

Abstract

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.