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 exclusive model access! 😉 - https://x.com/fluidvoiceapp

What it solves

FluidVoice is an open-source voice-to-text dictation app for macOS that allows users to write, rewrite, and control their computer using their voice. It addresses the need for a private, low-latency transcription service that can integrate directly into any application on the Mac without requiring cloud-based processing.

How it works

The app uses a variety of supported speech models (such as Nemotron, Parakeet, Cohere, and Whisper) to convert speech to text in real-time. It employs accessibility APIs to insert text directly into any active text field. It also features "Fluid Intelligence," a private, on-device AI runtime for smart formatting and context-aware capitalization.

Who it’s for

Mac users who want a fast, private, and integrated voice dictation and system control experience, ranging from those who prefer fully local processing to those who who want to use cloud AI providers for enhancement.

Highlights

  • Local-First Privacy: Voice and text data remain on the machine unless the cloud AI is opted into.
  • Command Mode: Allows users to launch apps, run shortcuts, and automate workflows via voice.
  • Write Mode: Enables dictating or rewriting text directly within any application.
  • Multiple Model Support: Supports a wide range of models including Nemotron Speech 3.5, Parakeet, and Whisper for different language and latency needs.
  • Fluid Intelligence: A local AI layer for on-device post-processing and smart formatting.
  • Notch-Aware Overlay: A real-time transcription preview that fits around the MacBook notch.

Related

  • Project
    FluidInference/FluidAudio

    FluidAudio is a Swift SDK for on-device audio AI on Apple devices — transcription, text-to-speech, voice activity detection, speaker diarization, and speech enhancement — all running on the Apple Neural Engine for speed and privacy.

  • Project
    Beingpax/VoiceInk

    VoiceInk is a macOS‑only, offline speech‑to‑text app that uses local Whisper‑based models (via whisper.cpp and transcribe‑cpp‑swift) to provide near‑instant, privacy‑preserving transcription, context‑aware modes, custom dictionaries, global shortcuts, and a built‑in AI assistant. It’s open‑source under GPL‑v3, installable via Homebrew or from source, with optional paid licensing for updates and support.

  • Project
    TypeWhisper/typewhisper-mac

    A system-wide speech-to-text and AI text processing utility for macOS that supports local and cloud transcription engines with customizable AI workflows.

  • Project
    mylxsw/typeflux

    A macOS application that provides lightning-fast voice-to-text injection into any app, supporting local models for privacy and an AI assistant for voice-driven Q&A and editing.

  • Project
    watzon/pindrop

    Pindrop is a discontinued, native‑macOS menu‑bar dictation app that transcribes speech offline using on‑device models (Whisper‑Kit, Parakeet, SenseVoice, Apple Speech) or optional OpenAI cloud models. It offers real‑time streaming, multilingual support, speaker diarization, a searchable transcript library, AI‑enhanced cleanup, and privacy‑first defaults. Built in Swift/SwiftUI, it runs on macOS 14+ (Apple Silicon recommended) and is replaced by the Sayso app.