T(ime) T(o) S(tart) synthesising audio description in China? | Results of a reception study

Image by Gerd Altmann from Pixabay

By Irene Tor-Carroggio

JoSTrans 34 (2020)

Permalink: https://jostrans.org/issue34/art_tor.php

Abstract | Text-to-speech audio description has proven to be an accepted method to increase the output of this access service in several languages, such as English, Japanese, Polish, Spanish and Catalan – at least as an interim solution until more audio description is available with human narrators. In the light of China’s need to make audio description more widely available, we propose text-to-speech audio description as a means to provide this access service both faster and more economically. This article describes and analyses the results of a reception study carried out in China to test the acceptance of text-to-speech audio description in comparison with standard human-voiced audio description. The study sample consisted of forty participants and was carried out using clips from a Chinese historical movie. With the exception of comprehension, the results show that natural voices score statistically higher than synthetic voices, which suggests that those with sight loss prefer human-voiced audio description. Yet, it cannot be stated that text-to-speech audio description is not accepted: more than half of the study participants not only see this as an interim alternative, but also as a welcome permanent solution if it means more audio described movies.