Skip to content

liaqateagle/WhisperDesk

Repository files navigation

WhisperDesk Logo

WhisperDesk

Free, open-source audio & video transcription. Local. Private. Beautiful.

Download · Features · SpeakDock (Pro Tool) · YouTube

License Platform Tauri v2 Stars


WhisperDesk Screenshot


Features

  • Local & Private — 100% offline transcription, no cloud, no API keys, no telemetry
  • Word-Level Timestamps — Click any word to jump to that point in the audio
  • Interactive Transcript — Real-time word highlighting during playback
  • Multiple Export Formats — SRT, VTT, TXT, JSON, Markdown
  • Model Manager — Download Whisper models (tiny → large-v3) with progress bars
  • Transcription History — SQLite-backed history with search and stats
  • Beautiful UI — Dark/light mode, glassmorphism, smooth animations
  • 99 Languages — Full Whisper language support
  • Built-in Audio Player — Play, pause, seek, speed control (0.5x–2x)

Supported Formats

Audio: MP3, WAV, M4A, OGG, FLAC, WMA, AAC

Tech Stack

  • Tauri v2 — Lightweight desktop framework (Rust + Web)
  • React 19 + TypeScript — Frontend
  • Tailwind CSS v4 — Styling
  • whisper-rs — OpenAI Whisper.cpp speech recognition
  • Symphonia — Pure Rust audio decoding
  • SQLite — Transcription history
  • Framer Motion — Animations
  • Lucide React — Icons

Getting Started

Prerequisites

Development

# Install dependencies
npm install

# Run in development mode
npm run tauri dev

# Build for production
npm run tauri build

First Use

  1. Go to Settings → Models and download a Whisper model (start with "base" for good quality/speed balance)
  2. Go to Transcribe and drop an audio file
  3. Click Start Transcription
  4. View, edit, search, and export your transcript

Whisper Models

Model Size Speed Quality
tiny 75 MB Fastest Basic
base 142 MB Fast Good
small 466 MB Medium Better
medium 1.5 GB Slow Great
large-v3 3.1 GB Slowest Best
large-v3-turbo 1.6 GB Fast Great

Part of the SpeakDock Ecosystem

WhisperDesk is the free, open-source companion to SpeakDock — our commercial real-time dictation and voice command tool.

WhisperDesk = File transcription (open source, free) SpeakDock = Real-time dictation, AI polish, meeting recording, voice commands (commercial)

Author

Liaqat Eagle

License

MIT — free for personal and commercial use. See LICENSE.

Contributing

Contributions welcome! See CONTRIBUTING.md for details.

About

Free, open-source audio transcription app. 100% local & private. Word-level timestamps, interactive transcript, multi-format export. Built with Tauri v2, React, and Whisper.cpp.

Resources

License

Contributing

Stars

6 stars

Watchers

0 watching

Forks

Packages

 
 
 

Contributors