The Team Behind the Engine: OpenAI & Faster-Whisper
Every time you speak into Vox Bar and watch your words appear on screen, you're using technology built by a small team in Paris who believed AI should belong to everyone. This is their story — and our thank you.
The Spark That Started It All: OpenAI's Whisper
In September 2022, OpenAI open-sourced Whisper, a general-purpose speech recognition model. Trained on a massive 680,000 hours of multilingual web audio, Whisper demonstrated unseen levels of robustness against accents, background noise, and technical jargon. It was a watershed moment: overnight, offline transcription matched the quality of premium cloud APIs.
By open-sourcing the model weights, OpenAI sparked a revolution in the open-source community, enabling developers worldwide to build private, local transcription tools. But there was one catch: the original Whisper implementation was slow and required heavy architectural resources.
The Open Source Optimizer: Guillaume Klein & SYSTRAN
Enter Guillaume Klein and the team at SYSTRAN. As experts in machine translation, SYSTRAN had built an incredibly efficient inference engine called CTranslate2. Guillaume adapted the Whisper architecture to run on CTranslate2, creating a project known as faster-whisper.
Faster-whisper was a revelation. It delivered up to 4x speedups compared to the original OpenAI codebase, while taking up significantly less VRAM. This meant that the accuracy of Whisper could now run on standard consumer laptops, not just powerful server racks.
The Backbone of Local AI
The combination of OpenAI's massive training efforts and Guillaume's optimization brilliance created the foundational engine for modern offline voice-to-text. In Vox Bar, our Starter (Whisper+) tier uses the faster-whisper engine to provide highly accurate, completely free transcription on nearly any hardware.
🌍
Merci to the pioneers
To the researchers at OpenAI, Guillaume Klein, and the CTranslate2 contributors — thank you for laying the foundation of the offline voice revolution.
Experience Faster-Whisper for yourself
Vox Bar brings Mistral's frontier transcription to your desktop. Private. Local. Yours.
Coming Soon Early Bird