Fact · Speech recognition
In the 2019 paper, a word counts toward word diarization error rate when it is a correctly recognized word with the wrong speaker label, or a substituted word with a wrong speaker label.
Checked against its source on October 17, 2026
In the source's own words
“CIS is the number of Correct ASR words with Incorrect Speaker tokens”
Joint Speech Recognition and Speaker Diarization via Sequence Transduction, Laurent El Shafey et al., Interspeech 2019, 2019, Section 5.2.5, equation 2.
Open the original- What it measures
- The paper defines word diarization error rate as a ratio built from these counts over the number of ASR substitutions and correct words.
- What it does not show
- The definition is from one paper. Other papers use related measures, such as concatenated minimum-permutation word error rate, so check which one a benchmark reports.