Average subtitling speed or how to compare apples and oranges

Subtitling speed (or captioning speed) is one of the most researched parameters of subtitled audiovisual materials. It is of paramount importance for pre-recorded programs and for live shows since fast subtitles that are not aligned with the viewers’ reading skills may not be read in full. If this occurs frequently, comprehension by the end users might be compromised. When it comes to research, the average subtitling speed is often reported as an indication of how fast (or how slow) the subtitles in a given program are delivered. Nevertheless, this measurement might not always reflect the actual speed characteristics of an audiovisual product.

To begin with, the same average subtitling speed may result from programs including subtitles delivered at very different paces. Additionally, there is no universal way to calculate this value and several mathematical formulas are used in the existing literature (as well as by professional subtitling software) to come up with what is always called “average subtitling speed”. Furthermore, depending on the specific characteristics of the audiovisual product, the use of one calculation method or another might lead to substantial variation in the results. This presentation will elaborate on the aforementioned ideas to describe why the notion of “average subtitling speed” is problematic and should be nuanced. Some reflections on the implications that this lack of uniformity might have for research will be shared with the audience, before reviewing potential alternatives to the use of the average subtitling speed per se.