A corpus-based analysis of audio description
Title of edited book
Media for All: Subtitling for the Deaf, Audio Description and Sign Language
Year of publication
This paper presents the beginning of a corpus-based investigation into the language used for audio description. The automated analysis of audio description scripts for 91 films was successful in characterising some idiosyncratic features of what appears to be a special language. Our investigation also began to create an empirically-grounded overview and classification of the main kinds of information provided by audio description. The existence of a special language is explained in part by the fact that audio description is produced by trained professionals following established guidelines, and its idiosyncrasies are explained by considering its communicative function – in particular that it is being used to tell a story. Encouraged by the relatively high degree of regularity observed in the corpus, we go on to speculate about the application of language technologies for ‘assisted audio description’ and for repurposing audio description as a basis for indexing digital video archives.