Glossary
whisper.cpp
whisper.cpp is a C/C++ port of OpenAI's Whisper speech recognition model, designed to run speech-to-text transcription locally on ordinary hardware.
What it means
Whisper is a speech recognition model OpenAI released openly. The original runs in Python with heavy dependencies. whisper.cpp is an independent re-implementation in C and C++ that runs the same models with no Python, no GPU requirement, and a small footprint.
That makes it a common building block for desktop apps that need transcription without a cloud service. It loads a model file once and then transcribes audio entirely offline, with optional GPU acceleration when one is available.
How this relates to Autorec
Autorec uses whisper.cpp for its built-in transcription. When a recording finishes, the audio is passed to whisper.cpp on your own machine, and the result is written out as a text transcript and an SRT subtitle file.
Try Autorec
A local-first meeting recorder for Linux and Windows. It auto-detects your calls, records to your own disk, and transcribes on your machine. One-time €20, with a free tier to start.
Download Autorec