amicalhq/amical

🎙️ AI Dictation App - Open Source and Local-first ⚡ Type 3x faster, no keyboard needed. 🆓 Powered by open source models, works offline, fast and accurate.

Amical – Local‑first AI dictation

What it is – Amical is an open‑source desktop application that lets you dictate text and take notes using AI, all on your own machine. It combines OpenAI’s Whisper for speech‑to‑text with locally‑run LLMs (via Ollama) to understand context and format the transcription for the app you’re currently using (email, IDE, chat, etc.). No internet connection is required, so your voice data never leaves the device.

Key features

  • Fast, accurate dictation – Whisper runs locally, giving near‑real‑time transcription.
  • Context‑aware formatting – The app detects the active window and adapts the output (e.g., email style, code snippets, chat messages).
  • Extensible workflow – Hotkeys, voice macros, and custom pipelines can be added.
  • Privacy‑first – All models are downloaded and executed locally; the app does not send audio to the cloud.
  • Floating widget – A small on‑screen control for start/stop without breaking your flow.
  • Planned – Summarized voice notes, task extraction, meeting transcription, and deeper app‑control integrations.

Technology stack

  • Speech‑to‑text: Whisper (compiled as a native addon)
  • LLM inference: Ollama (runs open‑source models locally)
  • Desktop framework: Electron + TypeScript
  • UI: TailwindCSS + Shadcn UI components
  • Auth & validation: Better‑Auth, Zod
  • Build & testing: Turborepo, Vitest, TypeScript 7 (native compiler)

Getting started (development)

  1. Clone with submodules – git clone --recurse-submodules https://github.com/amicalhq/amical.git
  2. Install tools – Node 24, pnpm 10.15, CMake ≥ 3.20, plus Xcode (macOS) or Visual Studio 2022 (Windows).
  3. Enable Corepack & install deps – corepack enable && pnpm install --frozen-lockfile
  4. Download the Node binary used by Whisper – pnpm --filter @amical/desktop download-node
  5. Run the app – pnpm turbo run dev --filter=@amical/desktop
  6. Run checks before PRs – pnpm type:check && pnpm test

Installation for end users

  • macOS: brew install --cask amical or download the latest release.
  • Windows: download the installer from the Releases page.
  • Android: available via Google Play (link in README).
  • iOS: beta sign‑up link provided.

Contributing – The project welcomes PRs and bug reports. Discussions and community support happen on their Discord server. Follow the contribution checklist (run type‑check and tests) before opening a PR.

License – MIT.

Related

  • Project
    OpenWhispr/openwhispr

    A privacy-first, open-source voice-to-text application providing local AI dictation, meeting transcription with speaker diarization, and AI-powered notes.

  • Project
    Notely-Voice/NotelyVoice

    A private, cross-platform AI voice transcription and note-taking app that uses OpenAI Whisper to provide offline speech-to-text capabilities.

  • Project
    akshayaggarwal99/jarvis-ai-assistant

    An open-source voice dictation tool that converts speech to text anywhere on your system, supporting local transcription and LLM post-processing for privacy and speed.

  • Project
    voquill/voquill

    An open-source, cross-platform AI voice typing app that enables dictation into any desktop application with AI-driven text cleanup and a personal glossary.

  • 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.