Towards research-informed training in interlingual respeaking: an empirical approach

Abstract

Respeaking is an effective way to make live and pre-recorded television, as well as live events, accessible to a wide audience. Respeakers use speech-recognition software to repeat or paraphrase what is heard through a microphone in a robotic voice while enunciating punctuation and adding colours or labels to identify the speakers. Until now, most respeaking has been done within the same language (intralingual respeaking). As a recent development, interlingual respeaking has the potential to provide a new access service for a wider audience, including those with hearing loss, language learners, foreigners, migrants, refugees, etc. To ensure quality, respeakers must receive systematic and appropriately designed training. As part of the EU-funded project Interlingual Live Subtitling for Access (ILSA), whose goal is to develop a training course for interlingual live subtitlers, this paper presents the results of the largest experiment on interlingual respeaking conducted so far. The aim was to identify the skills required for interlingual respeaking and determine the best-suited professional profile for an interlingual live subtitler to inform training. The results show a varied and complex landscape where subtitlers and interpreters must learn different skills to perform effectively as interlingual respeakers.