Effort in Semi-Automatized Subtitling Processes

Anke Tardel

doi:10.47476/jat.v3i2.2020.131

Abstract

The presented study investigates the impact of automatic speech recognition (ASR) and assisting scripts on effort during transcription and translation processes, two main subprocesses of interlingual subtitling. Applying keylogging and eye tracking, this study takes a first look at how the integration of ASR impacts these subprocesses. 12 professional subtitlers and 13 translation students were recorded performing two intralingual transcriptions and three translation tasks to evaluate the impact on temporal, technical, and cognitive effort, and split-attention. Measures include editing time, visit count and duration, insertions, and deletions. The main findings show that, in both tasks, ASR did not significantly impact task duration, but participants had fewer keystrokes, indicating less technical effort. Regarding visual attention the existence of an ASR script did not decrease the time spent replaying the video. The study also shows that students were less efficient in their typing and made more use of the ASR script. The results are discussed in context of the experiment and an outlook on further research is given.

Highlights

Audiovisual productions contribute tremendously to Europe’s economy and culture
Summarizing the explorative findings and the automatization potentials within subtitling processes, current state-of-the-art automatic speech recognition (ASR) seems not to come near the effects correct transcripts have in terms of temporal, technical and cognitive effort
Technical effort was reduced by the ASR as fewer keystrokes were performed and less time was spent typing

Summary

Introduction

Audiovisual productions contribute tremendously to Europe’s economy and culture. The main asset of the audiovisual market is the content of TV production companies. While the immediate market size of audiovisual content is limited by the production language, other markets can be reached by re-purposing the content via translation. Audiovisual translation (AVT) is mostly done via dubbing, voice-over, or subtitling. Dubbed versions are complex and expensive to produce, so a Europe-wide distribution of local audiovisual programmes cannot be achieved for all productions (Raats, Evens, & Ruelens, 2016). This depends on the distribution platform, and on the practices in individual countries (Baños & Diaz-Cintas, 2017)

Methods

Results

Conclusion