Fact · Dictation software
Researchers describe word error rate as a standard measure of the discrepancy between machine and human transcriptions.
Checked against its source on October 16, 2026
In the source's own words
“a standard measure of discrepancy between machine and human transcriptions”
Racial disparities in automated speech recognition, Allison Koenecke et al., Proceedings of the National Academy of Sciences, 2020, Methods, Word error rate.
Open the original (DOI 10.1073/pnas.1915768117)- What it measures
- A study of five commercial speech recognition systems, which uses word error rate to compare each system's transcripts with human transcripts.
- What it does not show
- The definition is the study's wording. The exact formula, which counts substitutions, deletions and insertions, is in the paper's methods.
Source
Cited on
- Dictation Software: How to Choose the Right AppTopic
- Speech Recognition: What It Is and How Well It WorksTopic
- Google Docs Voice Typing: How to Use It and Its LimitsGuide
- Best Dictation Software and Apps, Compared by NeedGuide
- Best Dictation Software for Windows: Free and PaidGuide
- Speech to Text for Deaf and Hard of Hearing PeopleGuide
- Verbatim vs Clean Verbatim TranscriptionGuide
- How do you test dictation software before buying?Question
- What does it mean to transcribe audio?Question