Glossary
On-device transcription
On-device transcription is speech-to-text conversion that runs entirely on your own computer, without uploading the audio to a remote server.
What it means
Cloud transcription sends your audio to a vendor's servers. Those servers need fast hardware and a network connection, but they also see everything you record. On-device transcription runs the speech recognition model on your own CPU or GPU instead, so the audio never leaves the machine.
The trade-off used to be quality and speed. That gap has narrowed: compact open speech models now run well on ordinary laptops, which makes on-device transcription practical for everyday meetings.
How this relates to Autorec
Autorec transcribes locally with whisper.cpp. The model files are downloaded once and stored on your machine. After that, transcription works offline and the audio of a sensitive call is never sent anywhere.
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