Skip to content

Releases: eddmann/VoiceScribe

VoiceScribe v1.0.1

26 Nov 10:30

Choose a tag to compare

VoiceScribe v1.0.1

Installation

  1. Download VoiceScribe-1.0.1.zip
  2. Unzip and move VoiceScribe.app to Applications
  3. Double-click to open

This release is signed and notarized by Apple.

Features

  • Global hotkey recording - Press Option-Shift-Space to record from anywhere (customizable)
  • Dual transcription engines - Choose between privacy-focused local WhisperKit (on-device, Apple Silicon only) or cloud-based OpenAI Transcription
  • Multiple AI models - Download and switch between WhisperKit models (Base/Small/Medium) or select OpenAI models (Whisper V2/GPT-4o/GPT-4o Mini)
  • AI-powered enhancement - Optional post-processing to add perfect punctuation, capitalization, and formatting
  • Smart paste - Transcriptions paste directly into your active app
  • Transcription history - Review past transcriptions anytime
  • Secure storage - API keys encrypted in macOS Keychain

Requirements

  • macOS 14.0 (Sonoma) or later
  • Apple Silicon (M-series) Mac for local WhisperKit transcription and MLX enhancement
  • Intel Macs supported with OpenAI Transcription API

Privacy

  • All audio processing happens on-device with WhisperKit (no network transmission)
  • OpenAI mode sends audio to their servers - review OpenAI's Privacy Policy
  • API keys encrypted in macOS Keychain with device-only access
  • No telemetry or analytics

VoiceScribe v1.0.0

12 Nov 09:05

Choose a tag to compare

VoiceScribe v1.0.0

Installation

  1. Download VoiceScribe-1.0.0.zip
  2. Unzip the file
  3. Move VoiceScribe.app to your Applications folder
  4. Right-click and select "Open" the first time (to bypass Gatekeeper)

If you see "VoiceScribe is damaged and can't be opened", run:

xattr -cr /Applications/VoiceScribe.app

Features

  • Global hotkey recording - Press Option-Shift-Space to record from anywhere (customizable)
  • Dual transcription engines - Choose between privacy-focused local WhisperKit (on-device, Apple Silicon only) or cloud-based OpenAI Transcription
  • Multiple AI models - Download and switch between WhisperKit models (Tiny/Base/Small/Medium) or select OpenAI models (Whisper V2/GPT-4o/GPT-4o Mini)
  • Auto-paste - Transcriptions paste directly into your active app
  • Transcription history - Review past transcriptions anytime
  • Secure storage - API keys encrypted in macOS Keychain

Requirements

  • macOS 14.0 (Sonoma) or later
  • Apple Silicon (M-series) Mac for local WhisperKit transcription
  • Intel Macs supported with OpenAI Transcription API

Notes

⚠️ Unsigned Build: This build is not notarized or signed with a Developer ID. You'll need to right-click and select "Open" the first time you run it.

For auto-paste functionality, grant Accessibility permissions when prompted in System Settings → Privacy & Security → Accessibility.

Privacy

  • All audio processing happens on-device with WhisperKit (no network transmission)
  • OpenAI mode sends audio to their servers - review OpenAI's Privacy Policy
  • API keys encrypted in macOS Keychain with device-only access
  • No telemetry or analytics