Overview
Whisper is an open-source speech recognition model and Python package from OpenAI. It supports multilingual transcription, speech translation into English, language identification, and voice activity detection. The repository includes model code, command-line usage, Python examples, model cards, and multiple model sizes with different speed and memory needs. Developers can install it as a package or run it locally with PyTorch and FFmpeg. The official repository contains the product’s source code and is released under the MIT License, so it can be used and modified in self-hosted projects subject to that license. It is intended for developers building transcription and audio-processing workflows.