Fact · Voice and AI
Mistral's Voxtral models are trained to understand both spoken audio and text documents, and the paper calls them multimodal audio chat models.
Checked against its source on October 17, 2026
In the source's own words
“Voxtral is trained to comprehend both spoken audio and text documents”
Voxtral, Mistral AI, arXiv (Audio and Speech Processing), 2025, Abstract.
Open the original- What it measures
- The abstract introduces Voxtral Mini and Voxtral Small, which are the two models compared in the paper's per-language tables.
- What it does not show
- This describes the authors' own model design. It is the paper's claim, and the per-language numbers in the same paper are its evidence.