Title
Audiosubtitling. Making multilingualism accesible. A study on user experience
Conference name
Barrier-free Communication: Methods and Products
City
Country
Switzerland
Modalities
Date
15/09/2017
Abstract
Access services have received a lot of attention in audiovisual translation (AVT) research recently. However, there are still some modalities such as audio subtitling (AST) that have not been analysed in depth. In dubbing countries such as Spain, multilingualism is often maintained in films by using subtitles for secondary languages. For certain audiences, though, such as the visually impaired or people with reading difficulties an audio version of the subtitles (i.e. audio subtitling) may help providing an entire understanding of the audiovisual contents. AST is a technique that drags characteristics from AVT strategies such as dubbing, voice-over or subtitling. This presentation aims at defining AST and explaining the particularities of this transfer mode in terms of applications (contents and supports) and treatment of multilingualism.
AST are differentiated according to the way they are applied and elaborated. The first main distinction should be made between the two broadcasting systems: broadcast or user mix, depending on where the audio tracks are merged. And secondly, depending on the type of voice used, synthetic or human. It is of high importance to be aware of such categories when considering their implementation and processing.
The present study focuses on pre-recorded (broadcast mix) human-voiced AST for fiction multilingual films, where there are different ways of delivering AST. By combining the original track and the AST track and providing different information about the languages spoken, multilingualism can be rendered more or less explicitly, in the same way dubbing or subtitling can reveal more or less the presence of the foreign. Part of this research aims to apply to AST the model of linguistic representation developed by Sternberg (1981) used by Szarkowska et al. (2013) in subtitling for the deaf and hard-of-hearing.
By means of an analysis of a corpus of films, the strategies (referred to as voice-over and dubbing effects) previously described only in a few works (Braun & Orero, 2010; Remael, 2014) have been identified as the most popular ways of AST delivery. Their differences can be found primarily in the audibility of the tracks, their isochrony and prosodic features of the utterances. Furthermore, even if such effects appear in previous works, their definition has not been consciously delimited yet, this research aims at providing a detailed description of their features.
AST are differentiated according to the way they are applied and elaborated. The first main distinction should be made between the two broadcasting systems: broadcast or user mix, depending on where the audio tracks are merged. And secondly, depending on the type of voice used, synthetic or human. It is of high importance to be aware of such categories when considering their implementation and processing.
The present study focuses on pre-recorded (broadcast mix) human-voiced AST for fiction multilingual films, where there are different ways of delivering AST. By combining the original track and the AST track and providing different information about the languages spoken, multilingualism can be rendered more or less explicitly, in the same way dubbing or subtitling can reveal more or less the presence of the foreign. Part of this research aims to apply to AST the model of linguistic representation developed by Sternberg (1981) used by Szarkowska et al. (2013) in subtitling for the deaf and hard-of-hearing.
By means of an analysis of a corpus of films, the strategies (referred to as voice-over and dubbing effects) previously described only in a few works (Braun & Orero, 2010; Remael, 2014) have been identified as the most popular ways of AST delivery. Their differences can be found primarily in the audibility of the tracks, their isochrony and prosodic features of the utterances. Furthermore, even if such effects appear in previous works, their definition has not been consciously delimited yet, this research aims at providing a detailed description of their features.