Voxtral Transcribes at the Speed of Sound: Introducing Voxtral Transcribe 2
We're thrilled to unveil Voxtral Transcribe 2, a groundbreaking leap in speech-to-text technology. This release introduces two cutting-edge models: Voxtral Mini Transcribe V2 and Voxtral Realtime, each designed to revolutionize transcription across diverse applications.
Voxtral Mini Transcribe V2: Unparalleled Transcription Quality
Mini Transcribe V2 sets a new benchmark in transcription accuracy. It boasts state-of-the-art performance with speaker diarization, context biasing, and word-level timestamps in 13 languages. This level of precision is ideal for meeting transcription, interview analysis, and multi-party call processing.
Voxtral Realtime: Real-Time Transcription with Unmatched Latency
Voxtral Realtime is purpose-built for live transcription applications. Its innovative streaming architecture enables transcriptions with latency as low as sub-200ms, making it perfect for voice agents and real-time applications.
Best-in-Class Efficiency and Open-Source Accessibility
Voxtral Mini Transcribe V2 delivers industry-leading accuracy at a fraction of the cost, achieving the lowest word error rate at its price point. Voxtral Realtime, available under the Apache 2.0 license, offers open-weights deployment on edge devices for privacy-first applications.
Multilingual Excellence and Real-World Applications
Both models excel in multilingual transcription, supporting 13 languages including English, Chinese, Hindi, Spanish, Arabic, French, Portuguese, Russian, German, Japanese, Korean, Italian, and Dutch. Voxtral Realtime, in particular, shines with a 4B parameter footprint, ensuring efficient performance on edge devices.
Transforming Voice Applications
Voxtral's technology empowers a wide range of voice applications:
- Meeting Intelligence: Transcribe multilingual recordings with precise speaker attribution, enabling efficient annotation of large meeting volumes.
- Voice Agents and Virtual Assistants: Build natural-sounding conversational AI with ultra-low latency transcription.
- Contact Center Automation: Real-time transcription for sentiment analysis, response suggestions, and CRM field population during calls.
- Media and Broadcast: Generate live multilingual subtitles with minimal latency, handling technical terminology with context biasing.
- Compliance and Documentation: Monitor and transcribe interactions for regulatory compliance, ensuring clear speaker attribution and audit trails.
Get Started with Voxtral Transcribe 2
Voxtral Mini Transcribe V2 is available now via API at $0.003 per minute. Voxtral Realtime is accessible via API at $0.006 per minute and as open weights on Hugging Face. Explore the documentation for detailed insights into Mistral's audio and transcription capabilities.
Join us in shaping the future of speech AI! Apply to join our team and be part of this exciting journey.
