Audio description is an intersemiotic translation in which images are translated into words. This access service allows users to understand and enjoy audiovisual content even when the visuals are not accessible to them. However, the provision of audio description is uneven across countries and research is relatively recent. A deeper understanding of how images are transferred into words is still needed, and especially relevant is the analysis of how audio description is approached by different describers in diverging cultural environments and also how end users receive the audio described content. One of the setbacks often encountered by researchers in the field of audio description is the lack of freely available comparable materials to use in their analysis and experiments, very often due to copyright reasons.

Inspired by Wallace Chafe’s Pear Stories Project, and by its application in audio description within the DTV4ALL project by Pilar Orero (see Mazur and Kruger), Visuals Into Words (VIW) has developed a multimodal and multilingual corpus of audio descriptions departing from a single stimulus, a short film created ad hoc for the project in English, and translated into other languages. The corpus allows studies to be carried out comparing the various audio descriptions produced for one specific language but also contrasting various languages.

The project is built upon two pillars: on the one hand, it has a strong open access component. All materials are freely available to the research community through this website. On the other, it aims to be a scalable and expanding project. Although limited in size in its first steps, it aims to involve external researchers and widen its scope.

This project has been funded by the scaled “Ministerio de Economía y Competividad”, reference code FFI2015-62522-ERC (October 2015-September 2016). The short film production has been co-funded by project FFI2012-39056-C02-01. The project is led by Anna Matamala, a member of TransMedia Catalonia, funded by the Catalan government (2014SGR027).