Fact · Languages and accents
27.1percent word error rate
On the Fleurs benchmark, the Whisper large-v2 model had a word error rate of 27.1 percent for Hebrew.
Checked against its source on October 17, 2026
In the source's own words
“Whisper large-v2 4.5 12.5 4.2 3.0 21.9 32.9 9.7 13.8 8.3 15.4 102.7 88.9 27.1 21.5”
Robust Speech Recognition via Large-Scale Weak Supervision (full text), Alec Radford et al., OpenAI (arXiv), 2022, Table 13 (Fleurs), column Hebrew.
Open the original (DOI 10.48550/arXiv.2212.04356)- What it measures
- Read from the table in the appendix of the Whisper paper. The row lists the 14 languages in the order of the table header, with Hebrew 13th.
- What it does not show
- Fleurs is read speech, not dictation. The paper's own measure is word error rate after the Whisper text normalizer, which is not identical to the measure in other tests.