karansinghgit/speaktype

100% offline, open-source, Voice Dictation app on any OS

What it solves

SpeakType is a privacy-focused voice-to-text tool that allows users to dictate text into any application on their computer without sending audio data to external servers. It eliminates the need for paid subscriptions and internet connectivity for transcription, providing a free, local-first alternative to cloud-based voice typing apps.

How it works

The application runs locally on macOS, Windows, and Linux. Users download a speech model (such as OpenAI Whisper or NVIDIA Parakeet) directly to their machine. By holding a customizable shortcut key, the user speaks, and the app transcribes the audio in real-time and inserts the text directly at the cursor position in any active text field.

Who it’s for

People who want a fast, free, and private way to type with their voice across all their desktop applications, especially those who prefer offline functionality and total control over their data.

Highlights

  • 100% Local Processing: All transcription happens on the user's computer, ensuring privacy and offline capability.
  • Broad Compatibility: Works across all apps and operating systems (macOS, Windows, Linux).
  • Hspports Multiple Models: Users can choose between different models (e.g., Parakeet for speed or Whisper for 99 languages).
  • Customizable Dictionary: Allows users to teach the app specific names and terms to improve accuracy.
  • Transcription History: Maintains a local history of dictations for easy copying and replaying.
  • Automatic Cleanup: Removes filler words like "um" automatically.

Related

  • Project
    HKUDS/CatchMe

    CatchMe is a local, privacy‑first tool that continuously records your computer activity (mouse, keyboard, windows, clipboard, screenshots, etc.), organizes it into a hierarchical activity tree, adds LLM‑generated summaries, and lets you query the whole history via CLI, a web dashboard, or an AI‑agent skill—without needing vector embeddings.

  • Project
    Firepal/stammer

    A Python utility that reconstructs one audio source using the most spectrally similar frames from another audio or video source.

  • Project
    prathoshap/vagdhenu

    A production-grade Sanskrit chant text-to-speech system that generates traditional melodic recitation instead of flat read-aloud speech.

  • Project
    duckbugio/flock

    Flock is a Docker‑based service that runs a Claude‑Code powered dev team you control via chat (Telegram, VK, or a text‑only adapter). From a single message it plans, writes, tests, reviews and opens a PR, with optional loops for self‑verification, goal evaluation, scheduled jobs and CI watching. Each chat gets its own isolated workspace and branch, and the system works with GitHub/GitLab/Gitea using a PAT. Deployment is a one‑line `docker compose up -d` after filling a small `.env` file; the core logic lives in `core/agents/` and adapters are thin wrappers.