Fuck You Flow
EL

Home / Local Dictation Privacy and Offline Operation

Local Dictation Privacy and Offline Operation

The default speech engine runs on your computer. This technical explanation distinguishes local recognition from updates and optional online services.

Luram AI Agency · Updated

Local speech is the default

FU Flow ships Whisper models and runs recognition through whisper.cpp on the same PC. No FU Flow account or paid API key is required for that mode. Once installed, local dictation can work without an internet connection.

An optional remote provider changes the data flow

Settings also offers an OpenAI-compatible speech provider. When you choose it, the app sends recorded audio to the endpoint you configured. The provider may require credentials and charge for processing. Its privacy and retention terms apply. Keep the local provider selected if you want speech recognition to stay on your PC.

Update checks use the network

The app checks for a newer release after startup. Checking does not download an installer. If you choose an update, the app downloads it. These requests are distinct from speech transcription. The site and download host receive normal request metadata when your computer connects to them.

History and the destination app

Transcripts, dictionary rules and statistics are stored locally in the app database. Text is inserted through the Windows clipboard and paste workflow. Anyone with access to your local files or clipboard may be able to read it. The program receiving the text can store or sync it under its own settings.

Practical checks

Inspectable source

FU Flow is MIT licensed. The public source includes the provider selection, local and remote speech adapters and the update check. This page describes application behavior, not an independent security audit or a compliance certification.

Download for Windows

FU Flow 0.9.3 beta · 1.7 GB · Windows 10 / 11 · MIT