A audiodescrição sob a perspectiva da semântica de frames: análise dos frames evocados pelo texto da audiodescrição e pelas imagens dinâmicas num curta-metragem

Abstract

This M.A. thesis presents an exploratory study affiliated to Audiovisual Translation Studies and drawing on Frame Semantics (FILLMORE, 1982). The study evaluates the frames evoked in the audio description text and in the dynamic images (video scenes) of the Brazilian short film Eu não quero voltar sozinho (I do not want to come back alone). The objective is to investigate the frames evoked by the audio description text and those evoked by the dynamic images. To that end, frames in the audio description text and in the dynamic images were annotated and the most evoked frames in each mode explored, followed by a survey of the most evoked frames in both modalities according to the groups being arranged by FrameNet Brasil, the Root Type. In addition, this study also evaluates if the audiodescription text meets the criteria proposed in the Guia para Produções Audiovisuais Acessíveis (Guide to Accessible Audiovisual Productions) (NAVES et al., 2016). Annotations were made using FrameNet Brasil platforms, Webtool for text annotation and Charon for dynamic image annotation. The analysis was carried out using spreadsheets based on the reports generated. The results show that both modes operate concomitantly in evoking and invoking frames. From the analyzed data, it was found that all elements of communicative modes contribute to the construction of meaning in a multimodal text. In addition to that, the frames showed adherence to the criteria proposed by the Guia para Produções Audiovisuais Acessíveis.