French company Mistral has announced the release of a new audio AI model, Voxtral, aimed at businesses and offering unique capabilities in speech processing.
Innovative Features of Voxtral
Voxtral is designed to deliver practical speech intelligence in real-world applications and outperforms Whisper large-v3, one of the top open-source audio transcription models.
Variants and Specifications of the Models
Voxtral is powered by the large language model Mistral Small 3.1 and supports multiple languages including English, French, Spanish, and more. It can transcribe up to 30 minutes of audio and understand up to 40 minutes, allowing users to engage in conversations and ask questions.
The Future of Audio AI from Mistral
Mistral has released two versions of the model: Voxtral Small with 24 billion parameters and Voxtral Mini with 3 billion parameters. Both options are available for download and local hosting, with integration into applications done via a single API call.
Voxtral from Mistral marks a significant advancement in the field of audio AI, providing high-quality speech processing and expanded functionalities for businesses.
