Mistral Releases Voxtral 3 Speech Models and LiveKit-Based Voice Agent SDK

4 reports 3 reporting sources 1 hours agoNew

Understand the event

AI overview

Mistral AI released Voxtral 3 transcription and speech generation models, a LiveKit-based voice agent SDK and a voice agents API, with the models available through its EU-hosted API or for deployment on customers’ infrastructure. Organizations seeking to run the full stack on-premises or in a private cloud must contact Mistral’s team. Transcription supports more than 50 languages, while speech generation supports more than 35 languages and more than 50 dialects. The SDK combines listening, reasoning, tool calling and speaking in a streaming pipeline that Mistral says completes conversational turns in under one second. Artificial Analysis ranked Voxtral Small Transcribe 3 third on its Speech to Text leaderboard with a 2.1% AA-WER Index. It reported final-transcript AA-WER of 3.3% at 960 ms and 3.6% at 480 ms for Mini Transcribe Realtime 3, compared with its predecessor’s 5.2%. Artificial Analysis also ranked Voxtral 3 TTS 12th in both its Provider Voice and Controlled Voice arenas and reported a price of $13 per million characters.

Generated by AI from reports · Updated 5 minutes ago

Latest development10/21 00:53 (UTC+8)

Mistral releases Voxtral 3 speech models, with Small Transcribe 3 ranking #3 Artificial Analysis · X

Developments

2 developments

10/21 00:53 (UTC+8) · 1 reports
Artificial Analysis Evaluates Mistral Voxtral 3 Speech Models
Artificial Analysis · X：Mistral releases Voxtral 3 speech models, with Small Transcribe 3 ranking #3
10/20 20:00 (UTC+8) · 3 reports
Mistral Introduces Voxtral 3 Speech Models and Voice Agent SDK
Mistral AI：Mistral AI releases Voxtral 3 speech models and a voice agent SDK

Report timeline

Follow the reports to explore different sides of the event.

10/21

00:53
Artificial Analysis · X
Mistral releases Voxtral 3 speech models, with Small Transcribe 3 ranking #3
Mistral released Voxtral 3 TTS and three Speech to Text models, all evaluated by Artificial Analysis. Voxtral Small Transcribe 3 ranks #3 with a 2.1% AA-WER Index, while Mini Transcribe Realtime 3 achieves 3.3% final-transcript AA-WER at 960 ms and 3.6% at 480 ms, down from its predecessor’s 5.2%. Voxtral 3 TTS ranks #12 in both the Provider Voice and Controlled Voice arenas and costs $13/1M characters.
00:18
Artificial Analysis · X
Mistral releases Voxtral 3 speech models with new transcription and speech generation rankings
Mistral has released new speech models! Voxtral Small Transcribe 3 takes the #3 spot on the Artificial Analysis Speech to Text leaderboard, and Voxtral 3 TTS enters the top 12 in both Text to Speech arenas. Artificial Analysis also identifies Voxtral 3 as a new Text to Speech model: https://t.co/TgV72vGDT4
00:00
Mistral AI · X
Mistral introduces Voxtral 3 and a new voice agents API
Introducing Voxtral 3: state-of-the-art Transcribe and Speech models powering Mistral’s new voice agents API. 🧵 https://t.co/YA9HNVfq0u

10/20

20:00
Mistral AIPicks
Mistral AI releases Voxtral 3 speech models and a voice agent SDK
Mistral AI released Voxtral 3, a family of transcription and speech generation models, alongside a voice agent SDK built on LiveKit. Transcription supports 50+ languages, while speech generation supports 35+ languages and 50+ dialects; the SDK combines listening, reasoning, tool calling, and speaking in a streaming pipeline that Mistral says completes turns in under one second. The models are available through Mistral’s API, with on-premises or private-cloud deployment available by contacting its team.

The same kind of briefing as the one above: a few stories worth reading. Free. Mail starts after you confirm, and every letter has a one-click unsubscribe. Read a past issue

Event heat trend

There is not enough continuous observation data to plot a trend yet.

Keep reading

Latest daily · Issue 27TorchNeuron brings native PyTorch support for Hugging Face models on AWS Trainium