Rev AI's open-source derived English Automatic Speech Recognition (ASR) model. Trained on 200,000 hours of human-transcribed English audio, known for high accuracy on English speech and allowing control over transcription verbatimicity.