What is OpenAI Whisper?

Short answer

Whisper is OpenAI's open-source automatic speech recognition (ASR) model. Released 2022, MIT-licensed, supports 99 languages, and is the default engine for most modern caption tools.

Detail

Whisper is a transformer-based ASR model open-sourced by OpenAI in September 2022. It was trained on 680,000 hours of multilingual speech data and ships with several model sizes (tiny, base, small, medium, large, and large-v3 / turbo for speed). Whisper handles 99 languages, robust accent and noise tolerance, and produces word-level timestamps. The MIT license means anyone can self-host it for free; it's the engine behind SoCaptions and most caption tools that don't roll their own ASR.