General Questions

What is Voxtral?

Voxtral is an open-source speech understanding AI platform built on Mistral AI's technology. It provides state-of-the-art speech recognition, transcription, and natural language understanding capabilities for audio content.

How does Voxtral differ from other speech recognition services?

Voxtral offers superior accuracy with lower word error rates, built-in Q&A capabilities without additional LLM chaining, multilingual support, and is completely open source under Apache 2.0 license. It's also more cost-effective than commercial alternatives.

What languages does Voxtral support?

Voxtral supports 8+ languages including English, Spanish, French, Portuguese, Hindi, German, Dutch, and Italian. The system automatically detects the language of the audio content.

Is Voxtral free to use?

Yes, Voxtral models are open source and free to use. You can download and deploy them locally without any licensing fees. API usage may have associated costs depending on your deployment method.

Technical Questions

What audio formats are supported?

Voxtral supports common audio formats including MP3, WAV, M4A, and FLAC. The maximum file size is 50MB for optimal processing performance.

How accurate is Voxtral's transcription?

Voxtral achieves state-of-the-art word error rates (WER) that outperform Whisper and other leading models. Accuracy varies by language and audio quality, but typically ranges from 95-99% for clear audio.

Can I deploy Voxtral locally?

Yes, all Voxtral models are open source and can be deployed locally. The Mini model is optimized for easy local deployment, while the Small model offers advanced features for more complex setups.

What are the system requirements for local deployment?

Requirements vary by model size. The Mini model can run on consumer hardware with 8GB RAM, while the Small model requires more resources. GPU acceleration is recommended for optimal performance.

API & Integration

How do I get started with Voxtral's API?

Visit our documentation page for detailed API guides, code examples, and integration tutorials. We provide SDKs for popular programming languages and frameworks.

What is the API rate limit?

Rate limits depend on your deployment method. For local deployments, limits are based on your hardware capabilities. Cloud API usage has fair use policies to prevent abuse.

Can I use Voxtral for real-time transcription?

Yes, Voxtral supports real-time transcription with minimal latency. The system is optimized for interactive applications and live audio processing.

How does Voxtral's Q&A feature work?

Voxtral can answer questions directly about audio content without requiring additional language model chaining. It understands context and provides relevant answers based on the transcribed content.

Business & Licensing

What license does Voxtral use?

Voxtral is released under the Apache 2.0 license, which allows commercial use, modification, and distribution with minimal restrictions. Attribution is required in most cases.

Can I use Voxtral in commercial applications?

Yes, the Apache 2.0 license allows commercial use. You can integrate Voxtral into your products and services without licensing fees.

Do you offer enterprise support?

Yes, we offer enterprise support packages including custom deployments, training, and dedicated technical assistance. Contact our business team for more information.

How much does Voxtral cost?

The models themselves are free. Costs depend on your deployment method - local deployment has no ongoing costs, while cloud API usage may have associated infrastructure costs.

Support & Community

Where can I get help with Voxtral?

We provide multiple support channels: documentation, community forums, GitHub issues, and direct email support. Visit our contact page for all support options.

How can I contribute to Voxtral?

Contributions are welcome! You can contribute code, report bugs, suggest features, or help improve documentation. Visit our GitHub repository to get started.

Is there a community forum or Discord?

Yes, we have an active community on Discord where users can share experiences, ask questions, and collaborate on projects. Links are available on our community page.

How do I report bugs or request features?

Bugs and feature requests can be reported through GitHub issues or by contacting our support team. Please include detailed information about the issue or request.

Privacy & Security

How does Voxtral handle privacy?

Voxtral processes audio data to provide transcription services. For local deployments, all data stays on your infrastructure. Cloud API usage follows our privacy policy with appropriate safeguards.

Is my audio data stored?

For local deployments, no data leaves your infrastructure. For cloud API usage, data retention policies are outlined in our privacy policy. You can request data deletion at any time.

Is Voxtral GDPR compliant?

Yes, Voxtral is designed with privacy regulations in mind. Local deployments give you full control over data, while cloud services include GDPR-compliant data handling practices.

Can I use Voxtral for sensitive content?

Yes, local deployment ensures your sensitive audio content never leaves your infrastructure. This makes Voxtral suitable for healthcare, legal, and other privacy-sensitive applications.

Frequently Asked Questions