altic-dev/FluidVoice

Fastest and only macOS Dictation app with on-device STT and custom trained AI enhancement model. Windows pre-build available! A local Wispr Flow alternative. DM us on X for an easter egg 😉 - https://x.com/fluidvoiceapp

What it solves

FluidVoice is an open-source voice-to-text dictation application for macOS that allows users to type and control their computer using their voice. It eliminates the need for manual typing in any text field and provides a way to automate system actions and workflows through voice commands.

How it works

The app uses a variety of speech-to-text models (such as Nemotron, Parakeet, and Whisper) to transcribe audio in real-time. It then uses accessibility APIs to insert that text directly into any active application. For enhanced quality, it offers "Fluid Intelligence," a local AI runtime that handles smart formatting, capitalization, and post-processing on-device, or optional cloud-based AI providers (like OpenAI or Groq) for further refinement.

Who it’s for

It is designed for macOS users (specifically those on macOS 15.0+ with Apple Silicon, though Intel is supported via Whisper) who want a private, local-first dictation tool for writing, rewriting text, or controlling their Mac via voice.

Highlights

  • Multiple Model Support: Choice of various engines including Nemotron Speech 3.5, Parakeet, Cohere, and Whisper for different latency and language needs.
  • Command Mode: Ability to launch apps, run shortcuts, and trigger system actions via voice.
  • Write Mode: Direct insertion or rewriting of text in any app's text field.
  • Local-First Privacy: On-device processing ensures voice and text data stay on the machine unless cloud providers are explicitly enabled.
  • Live Preview: A notch-aware real-time transcription overlay that shows words as they are spoken.

Related

  • Project
    mylxsw/typeflux

    A macOS app that provides lightning-fast voice-to-text injection into any application, supporting both local and cloud-based AI models for dictation and AI assistance.

  • Project
    watzon/pindrop

    An open-source, Mac-native AI dictation app that converts speech to text using local Apple Silicon engines or optional cloud services for maximum privacy.

  • Project
    Beingpax/VoiceInk

    A native macOS application that uses local AI models to provide instant, privacy-focused voice-to-text transcription.

  • Project
    moinulmoin/voicetypr

    An offline-first voice-to-text dictation app for macOS and Windows that transcribes speech and inserts it directly into any active application.

  • Project
    tornikegomareli/Talkify

    A privacy-focused voice dictation and text-to-speech app for macOS that uses on-device processing and dedicated keys for instant multi-language switching.