Whisper
Open-source speech recognition model for transcription and translation across 99+ languages
Notes
Whisper is released free and open-source under the MIT License by OpenAI. It runs locally via pip install openai-whisper and provides speech recognition across 99+ languages, speech translation to English, language identification, and voice activity detection. Multiple model sizes (tiny, base, small, medium, large, turbo) are available; the turbo model is optimized for speed but is not trained for translation. Local hardware is required; OpenAI also offers Whisper as a paid hosted API.