Towards a methodology for voice quality analysis and character coherence in dubbing

Abstract

This paper presents a methodology for voice quality analysis in dubbing and other audiovisual translation modes that include revoicing. The degree of correspondence between source and target voices, namely character coherence, is of utmost importance in audiovisual translation if we are to deliver a faithful translation and avoid unnecessary stereotypical cliches that could lead the audience to biased narrative. Although a few attempts at analysing voice quality have been done in audiovisual translation (Bosseaux 2015, 2018, Rodríguez Fernández-Peña 2020), we believe the methodologies proposed are incomplete since fundamental elements, such as the phonation types, voice group and the fundamental frequency range are not included. Here, we will present a methodology that complements the ones suggested by the scholarly tradition and which offers a comprehensive analysis of the actors’ voice qualities resorting to the speech analysis software Praat, and which is explained in the analysis of Hollywood voices offered at the end. Voice quality and identity constructions will be considered taking into account disciplines such as psychology, phonetics and phonology, audiology, and musicology. In addition, we will present the different types of phonation according to the most relevant researchers, which, together with other physical elements, are fundamental in order to analyse voice quality and their communicative value.