Fact · Languages and accents
14.3percent word error rate
On the Fleurs benchmark, the Whisper large-v2 model had a word error rate of 14.3 percent for Korean.
Checked against its source on October 17, 2026
In the source's own words
“Model Croatian Hungarian Armenian Indonesian Icelandic Italian Japanese Javanese Georgian Kazakh Khmer Kannada Korean Luxembourgish ... Whisper large-v2 13.4 17.0 44.6 7.1 38.2 4.0 5.3 nan 105.0 37.7 99.7 37.0 14.3 88.0”
Robust Speech Recognition via Large-Scale Weak Supervision (full text), Alec Radford et al., OpenAI (arXiv), 2022, Table 13 (Fleurs), row Whisper large-v2, column Korean.
Open the original (DOI 10.48550/arXiv.2212.04356)- What it measures
- Table 13 is the paper's Fleurs error table, and Korean is scored by word error rate because Korean is not on the paper's list of languages scored by characters. Korean is one of the 102 Fleurs languages.
- What it does not show
- Read speech in good audio, from one benchmark. Korean writes spaces between words inconsistently, which can move word counts, so the figure is not a precise measure of spacing accuracy. Everyday Korean speech will score higher.