FU Flow icon
FU Flow icon

FU Flow

Free offline dictation for Windows that types what you say into any application, using a Whisper speech model that runs on your own machine.

Main window: the dictation key, the ready state and a live log of what was typed.

Cost / License

  • Free
  • Open Source (MIT)

Application type

Platforms

  • Windows  Windows 10 and 11, 64 bit. Around 4 GB of free disk space for the application and its bundled speech models.
Local history: every phrase dictated, kept on the machine and reusable.
Personal dictionary: names and terms the model hears wrong are corrected once and stay fixed.
+2
Statistics: words dictated, minutes saved and measured recognition times on this machine.
FU Flow alternatives  
1like
51alternatives
0articles

Features

Properties

  1.  Privacy focused

Features

  1.  No Tracking
  2.  No registration required
  3.  Works Offline
  4.  Dark Mode
  5.  Ad-free
  6.  Voice dictation
  7.  Voice recognition
  8.  Dictation
  9.  Speech to text

 Tags

FU Flow News & Activities

Highlights All activities

Recent activities

FU Flow information

AlternativeTo Categories

Audio & Music, AI Tools & Services

GitHub repository

  •  0 Stars
  •  0 Forks
  •  1 Open Issues
  •   Updated  
View on GitHub

Popular alternatives

View all
FU Flow was added to AlternativeTo by Lu-Ram on and this page was last updated .
No comments or reviews, maybe you want to be first?

Official Links

AppStores & Other Links

What is FU Flow?

FU Flow is a free, MIT-licensed dictation application for Windows 10 and 11. You hold the right Alt key, speak, release it, and the text appears in whatever program the cursor sits in: a browser, Word, an email, a chat window or a code editor. A hands-free mode is available for longer passages.

Speech recognition runs locally through whisper.cpp with Whisper models included in the installer, so dictation keeps working without an internet connection. There is no account, no API key, no weekly word limit and no subscription. GPU acceleration uses Vulkan where the hardware supports it, with a CPU path for machines without a usable card.

Thirty-three languages are available in Settings, with Greek and English the most tested. The app also carries a personal dictionary that learns names and terms it gets wrong, reusable snippets, local dictation history and a statistics card showing how much has been dictated. Dark and light themes are included, and the interface is available in English and Greek.

An optional setting can route speech to a remote OpenAI-compatible provider, in which case that provider receives the audio and sets its own fees. The local engine remains the default.

FU Flow is developed by Luram AI Agency in Thessaloniki, Greece. Version 0.9.8 is a beta under active development, and the full source is published on GitHub.