Fact · Languages and accents
13languages
Mistral says Voxtral Realtime is natively multilingual and transcribes in 13 languages: English, Chinese, Hindi, Spanish, Arabic, French, Portuguese, Russian, German, Japanese, Korean, Italian and Dutch.
Quote found word for word on the live source, last checked October 19, 2026
In the source's own words
“The model is natively multilingual, achieving strong transcription performance in 13 languages, including English, Chinese, Hindi, Spanish, Arabic, French, Portuguese, Russian, German, Japanese, Korean, Italian, and Dutch”
Voxtral transcribes at the speed of sound, Mistral AI, Mistral AI, 2026, Voxtral Realtime section.
Open the original- What it measures
- Voxtral Realtime is released as open weights under the Apache 2.0 license. The same announcement says that Voxtral Mini Transcribe V2 'now supports 13 languages' like Realtime.
- What it does not show
- Strong performance in a language is not the same as equal accuracy in each. Mistral's own FLEURS charts show different word error rates by language.
In the data
A row of How Many Languages Speech Recognition Supports, as Voxtral Realtime: 13 languages.
- System
- Voxtral Realtime
- Maker
- Mistral AI
- Kind
- Open model
- What is counted
- Languages
- On device
- Yes
- As of
- 2026-10
- Note
- Counted the 13 names in the quoted list. The same page says the model has a 4B parameter footprint and runs efficiently on edge devices, which is the basis for On device.
Source
Used to calculate
- In the languages supported by speech recognition systems dataset, the 10 published figures for “Open model” have a median of 62 languages, ranging from 13 to 1,600.
- The languages supported by speech recognition systems dataset holds 45 data points, each a published figure traced to its source.
- The languages supported by speech recognition systems dataset draws on 39 distinct sources.