Illustration of what FluidVoice does: Open-source macOS dictation app that transcribes speech on your Mac

altic-dev /

FluidVoice

Open-source macOS dictation app that transcribes speech on your Mac

FluidVoice lets you press a global hotkey, speak, and have the text typed into whatever app you are using. Speech recognition runs on your Mac, so your voice does not leave the machine unless you choose a cloud AI provider for cleanup.

You pick a speech model during onboarding. The choices include Nemotron, Parakeet, Cohere Transcribe, Apple Speech and Whisper, which trade off language coverage, download size and latency. Text goes in through macOS accessibility APIs, which is why the app asks for microphone and accessibility permissions.

On top of plain dictation there is optional AI enhancement for formatting and capitalization. One caveat: the local enhancement model, called Fluid Intelligence, is a separate runtime that the maintainers keep private. The app itself is GPLv3. The README presents it as a local alternative to paid dictation apps such as Wispr Flow, though that comparison comes from the repo description.

What it does

  • Several on-device speech models

    Choose from Nemotron, Parakeet, Cohere Transcribe, Apple Speech and Whisper. Whisper covers up to 99 languages and also runs on Intel Macs. The other models need Apple Silicon.

  • Optional AI enhancement

    Optional cleanup through OpenAI, Groq, a custom provider, or the local Fluid Intelligence model. API keys are kept in the macOS Keychain.

  • Write Mode and Command Mode

    Write Mode dictates or rewrites text in any text field. Command Mode controls the Mac by voice, such as launching apps and running shortcuts.

  • Live preview and per-app prompts

    A live transcription overlay shows words as you speak and fits around the MacBook notch. You can assign different prompt sets to different apps.

  • Audio history and stats

    Optional local audio history with budget controls and ZIP export, plus daily usage stats.

What you could build

  • Dictate emails, notes and chat messages into any Mac app with a hotkey.
  • Rewrite selected text in place using Write Mode.
  • Open apps and trigger shortcuts by voice with Command Mode.
  • Transcribe in one of about 40 languages with a local speech model, without sending audio to the cloud.
  • Compare Parakeet, Nemotron and Whisper models for speed and accuracy on your own Mac.

Get started

Install with Homebrew or download the latest release, then grant microphone and accessibility access. Set a global hotkey and pick a voice model in onboarding. It needs macOS 15 or later.

brew install --cask fluidvoice

Commands are copied from the project's README. Read it on GitHub for the full setup.

More in Productivity & personal

    ↑↓ to move, Enter to open, Esc to close