Autrans understands Tshivenḓa (Venda), an official language of South Africa, the way it is actually spoken, and keeps its diacritic letters.
Tshivenḓa is one of South Africa's official languages, spoken mainly in Limpopo. Its diacritic letters (ḓ, ṱ, ḽ, ṅ, ṋ) are distinct letters, not styling, and generic tools strip them. Autrans transcribes Tshivenḓa preserving those letters and its noun-class prefixes, without normalizing it into English. Add it under Settings, Language.
Word error rate 74.0% (95% confidence interval 70.3–77.7%) on read speech, NCHLT test set, measured Sep 2026.
Measured on 250 recordings. Word error rate is the share of words that differ from a human-written transcript, so lower is better. Read speech means people reading set sentences aloud, so conversation, background noise and phone-quality audio will usually score worse.
Test set: NCHLT speech corpus (SADiLaR).
Autrans handles standard Tshivenḓa. Treat the output as a draft to review, especially for regional speech.
Yes. Letters like ḓ, ṱ, ḽ, ṅ, and ṋ are distinct in Tshivenḓa and are preserved in the transcript.
Add it under Settings, Language. It is not switched on by default in South Africa, where the default holds the five most widely spoken languages that do not clash with each other; Tshivenḓa comes after them.
Send Autrans into your next meeting, use it in the room, or upload a recording. Your first 30 minutes are free, given once, and no credit card is needed to start.