OpenAI
Open-source speech recognition model with near-human accuracy across 99 languages
Whisper is OpenAI's open-source automatic speech recognition model, trained on 680,000 hours of multilingual audio. It transcribes and translates speech across 99 languages with near-human accuracy, even in noisy environments. Available for free local inference or via the OpenAI API at $0.006 per minute.
Open-source ASR with near-human accuracy across 99 languages — free to self-host.
Free (open-source, self-hosted) · OpenAI API $0.006/minute