Image Recognition and Content Comprehension in Media Accessibility: A Study on the Reception of Audiences with Limited Hearing

Abstract

This experimental study examined techniques of subtitling for the deaf and hard-of-hearing (SDH) from the point of viewers’ content comprehension and image recognition. Three groups of deaf and hard of hearing people (n = 78) participated in the study. The research focused on four parameters: speed of subtitles, speaker identification, emotion and sound indication. By selecting three usual techniques for each parameter, three subtitled versions of a movie were produced. Each group of participants watched one of the subtitled versions and comprehension and freeze-frame tests were employed to examine participants’ reception. Based on the results, except for the image recognition, there were significant differences among the techniques in terms of participants’ content comprehension. As for the techniques of speaker identification, color-coding caused significantly better content comprehension scores than tagging. Concerning sound comprehension, the three techniques were not significantly different from one another in terms of viewers’ content comprehension. Regarding emotion comprehension, the absence of information and emoticons helped the participants comprehend characters’ tone and mood better than emotion descriptions.

Keywords

• media
• subtitling for the deaf and hard of hearing
• content comprehension
• image recognition
• reception