|
Whisper is an automatic speech recognition (ASR) system by OpenAI, designed to transcribe spoken language into written text. Trained on large datasets of diverse audio, Whisper models generalise to many datasets and domains without the need for fine-tuning.
whisper.cpp is a lightweight C/C++ reimplementation of OpenAI’s Whisper, developed by Georgi Gerganov and many other contributors. As of March 2026, it supports multilingual speech recognition (99 languages), as well as language identification. It is available in Praat (see Acknowledgments).
© Anastasia Shchupak 2026-03-15