Fact · Speech recognition
The 2019 paper says word diarization error rate must be used together with word error rate, because it does not account for deletions and insertions.
Checked against its source on October 17, 2026
In the source's own words
“Note that this WDER metric must be used in combination with the ASR Word Error Rate (WER) to account for deletions and insertions”
Joint Speech Recognition and Speaker Diarization via Sequence Transduction, Laurent El Shafey et al., Interspeech 2019, 2019, Section 5.2.5.
Open the original- What it measures
- A system can have a low WDER by dropping hard words, so the paper pairs the two measures.
- What it does not show
- Read the two numbers together. A low speaker error rate alone does not show the transcript is complete.