Word Error Rate

A standard metric for evaluating speech recognition and captioning accuracy, calculated as the number of insertions, deletions, and substitutions needed to transform the transcribed text into the reference text, divided by the total number of words in the reference. Lower WER indicates better accuracy. While widely used in research, WER has limitations for accessibility evaluation as it treats all errors equally regardless of their impact on comprehension — a misrecognized function word may matter less than a misrecognized technical term.