Voxtral 3

Available via Mistral’s EU-hosted API or self-deployment, it ensures data sovereignty while delivering low-latency, human-like interactions for contact centers, virtual assistants, and hands-free copilots.Speech transcription and voice generation have come a long way, but building end-to-end voice agents that can listen, reason, interact, and handle complex workflows remains difficult. A reliable voice agent needs to do four things very well (and simultaneously): Listen accurately across languages, accents, speakers, and interruptions,Reason deeply about what users mean in specific contexts,Speak naturally with emotion and expression, and Take action over multiple steps through connected systems. Getting all four to work together reliably in real time remains a challenge. To address this, we introduce Voxtral 3: a new family of state-of-the-art speech-to-text and text-to-speech models, alongside a voice agent SDK that brings listening, thinking, tool calling, and speaking into a single real-time streaming pipeline.Voxtral 3 is built to create natural conversational experiences, from contact-center agents and service desks to virtual assistants and hands-free copilots. With support for dozens of languages, expressive speech, complex reasoning, and deep integration with your own tools and data, agents built with Voxtral can interpret nuanced intent, reason through sophisticated requests, and execute multi-step actions across connected systems.As with other Mistral models, the Voxtral 3 family is designed for sovereign deployment. Available today through Mistral's EU-hosted API or for self-deployment on your own infrastructure, it gives organizations control over where audio is processed, how data is handled, and whether conversations are retained.HighlightsAccurate transcription across 50+ languages. The speech-to-text model, Voxtral Small Transcribe 3, is state-of-the-art in multilingual transcription, identifying who said what (diarization), using keyphrases to improve its accuracy (context-biasing). It detects when someone has finished speaking so agents can respond without delay (end-of-turn detection).Natural speech that adapts to the conversation. The text-to-speech model, Voxtral Mini TTS 3, introduces new voices that adjust their emotion, pace, and emphasis based on conversational context, with support for 35+ languages and 50+ dialects.Human-like intelligence at your (spoken) command. The new voice agent SDK combines listening, reasoning, tool calling, and speaking into a real-time experience. Agents can retrieve information, interact with business systems, and complete tasks during live conversations over phone or web.Pro Voice for distinctly accurate voice adaptation. For professional use cases where basic voice cloning does not suffice, Voxtral’s Pro Voice adaptation is able to reproduce a speaker’s distinctive character, emotional range, and speaking style across languages, using just 30 minutes of audio. Your voice AI, on your infrastructure. Use Mistral's EU-hosted API or deploy Voxtral 3 on your own infrastructure. Keep sensitive conversations within your environment and control whether any audio is retained. Try Voxtral now  | Learn moreIt hears 50+ languages, with state-of-the-art accuracy.Voxtral Transcribe 3 represents a significant milestone across transcription, diarization (who spoke when) and context biasing capabilities. The supported list of languages is expanded to 50, including 24 official EU languages and many low-resource Asian and Indic languages.The Small version achieves state-of-the-art results across diarization and multilingual transcription, including in English, French and Spanish. It outperforms Whisper, Scribe v2 and Gemini 3.5 Transcribe. The Mini variant is 4B parameters, enabling fast and accurate transcriptions at a price of $0.0035/min.Both models support context biasing, where key terms and free-form text can be given to bias the transcription and improve recall on technical terms. This enables them to be used in production settings with domain-specific vocabulary without the need for fine-tuning.FLEURS 45 Forced vs InferredWord Error Rate — lower is betterForcedInferred4.1%4.0%4.1%4.0%4.6%4.5%6.1%4.5%4.9%4.9%6.3%5.6%16.0%14.9%Voxtral Small Transcribe 3Microsoft AI Transcribe 2Voxtral Mini Transcribe 3Gemini 3.5 TranscribeScribe v2GPT TranscribeWhisper large-v3Word Error RateMacro-average WER over the intersection of 45 languages in the FLEURS benchmark supported by all seven models. WER is used for all languages except Japanese, Korean, Thai, and Chinese, which use CER. Lower is better.Forced = language specified; Inferred = language auto-detected. Languages: AR, BN, BG, CA, ZH, CS, DA, NL, EN, ET, FI, FR, DE, EL, GU, HI, HU, IS, ID, IT, JA, KN, KO, LV, LT, MK, MS, ML, MR, NO, FA, PL, PT, RO, RU, SK, SL, ES, SV, FIL, TE, TH, TR, UK, VIDiarizationConcatenated Minimum Permutation WER — lower is better18.4%18.6%19.0%20.3%23.0%30.9%38.4%Voxtral Small Transcribe 3Voxtral Mini Transcribe 3Pyannote Orchestrator (Precision 3)Microsoft AI Transcribe 2Scribe v2Voxtral Mini Transcribe 2Gemini 3.5 TranscribeConcatenated Minimum Permutation WERMacro-average concatenated minimum permutation WER (cpWER) over public diarization benchmarks: AMI-IHM, AMI-SDM, Switchboard, Switchboard CallHome, AISHELL-4. For maximum compatibility of API baselines, we use 30 minute versions of AMI and AISHELL. Lower is better.Diarization benchmark, Concatenated Minimum Permutation Word Error Rate (cpWER).Keyphrase RecallWith and without contextual prompting — higher is betterWithout ContextWith Context94.6%81.4%94.3%84.0%87.3%80.9%Voxtral Small Transcribe 3Scribe v2Gemini 3.5 TranscribeKeyphrase RecallVoxtral Small Transcribe 3 is highlighted in brand colours. Keyphrase recall on the Earnings 22 benchmarks. The context bias contains keyterms and a randomly sampled number of distractors, ranging from 0 to 100. Bars are superposed absolute values — the outer bar is recall with context, the inner bar is recall without context; the visible band is the uplift from contextual prompting.

Higher is better.Keyphrase recall benchmark.