Captain's Log

An 80s-inspired voice note dictation and transcription application that provides a unique, efficient, and enjoyable user experience for capturing and managing spoken thoughts, ideas, and notes.

Features

• Segmented Voice Recording

  • Start, pause, and resume recording with a single cycle button
  • Each pause sends the current audio segment for transcription immediately
  • Resume works instantly — transcription runs asynchronously in the background
  • Visual feedback with retro-style VU meter (freezes during pause)
  • Spacebar hold-to-record support
  • Countdown timer with pause-aware elapsed-time tracking

• Groq Whisper Transcription

  • Automatic voice-to-text via Groq's whisper-large-v3-turbo API
  • Per-segment transcription with timestamp-ordered results
  • Confidence scores derived from segment log-probabilities
  • Support for multiple languages (en, es, fr, de)
  • Configurable model selection (turbo vs. accurate)

• Note Management

  • Create, edit, and delete log entries
  • Folder organization and tagging system
  • Search with keyword highlighting
  • Note detail modal with full transcription view
  • Edit notes with a textarea (preserves original segments)

• Audio Persistence & Playback

  • Recorded segments stored in IndexedDB for playback
  • Audio playback directly within the note detail modal
  • Configurable audio persistence toggle in settings

• Export & Sharing

  • Export notes as .txt files
  • Copy note text to clipboard with retro "TRANSMIT" animation

• Retro UI/UX

  • Authentic 80s computer aesthetic with CRT effects and scanlines
  • Green, amber, and white terminal color schemes
  • Period-appropriate sound effects (toggleable)
  • Adjustable font size (14–24px)
  • Classic terminal-style interface with VT323 monospace font

• Privacy

  • Local-first: all data stored in localStorage and IndexedDB
  • No cloud sync — your recordings stay on your device
  • API key stored locally in .env

Getting Started

Prerequisites

• Node.js (v18 or higher)
• A Groq API key for speech-to-text
• Modern web browser (Chrome, Firefox, Safari, Edge)

Installation

1. Clone the repository:

git clone https://github.com/technicalmonk/captains-log.git
cd captains-log

2. Install dependencies:

npm install

3. Create a .env file with your Groq API key:

VITE_GROQ_API_KEY=your_groq_api_key_here

4. Start the development server:

npm run dev

The application will open in your default browser at http://localhost:5173.

Available Scripts

• npm run dev — Runs the app in development mode with Vite
• npm run build — Type-checks and builds the app for production
• npm run preview — Previews the production build locally
• npm test — Runs the test suite (Vitest)
• npm run test:watch — Runs tests in watch mode

Technology Stack

• Frontend: React 19
• Build Tool: Vite 7
• Speech-to-Text: Groq Whisper API (whisper-large-v3-turbo)
• Audio Capture: MediaRecorder API
• Audio Storage: IndexedDB
• Data Storage: localStorage
• Styling: CSS Modules with CRT/retro effects
• State Management: React Hooks
• Testing: Vitest + Testing Library

Configuration

Settings are accessible via the CONFIGURE view in the app header:

Setting	Options	Description
Language	en, es, fr, de	Transcription language sent to Groq
Model	whisper-large-v3-turbo, whisper-large-v3	Speed vs. accuracy tradeoff
Sound Effects	ON / OFF	Toggle retro beep/startup sounds
Font Size	14–24px	Adjustable base font size
Audio Persistence	ON / OFF	Store recorded segments in IndexedDB for playback

Settings persist in localStorage and apply immediately.

Recording Flow

START RECORDING → click → recording begins, button becomes PAUSE
PAUSE → click → stops current segment, sends audio to Groq,
         transcription appears when ready, button becomes RESUME
RESUME → click → starts new segment, button becomes PAUSE
SAVE LOG → if recording: sends final segment, saves all results as a note
CLEAR LOG → stops recording, discards all transcription and audio
Timer expiry / spacebar release → sends final segment, stops recording

Browser Compatibility

Tested and supported in:

• Chrome (latest 3 versions)
• Firefox (latest 3 versions)
• Safari (latest 2 versions)
• Edge (latest 3 versions)

Note: MediaRecorder and IndexedDB are required. All modern browsers support these APIs.

Troubleshooting

Common Issues

1. Microphone Access

   • Ensure browser has microphone permissions
   • Check system microphone settings
   • Verify HTTPS if running in production (required for getUserMedia)

2. Transcription Issues

   • Confirm VITE_GROQ_API_KEY is set in your .env file
   • Check internet connection (Groq API requires network access)
   • Very short segments (<10s) are still billed at the 10s minimum by Groq

3. Build Problems

   • Delete node_modules and reinstall: rm -rf node_modules && npm install
   • Verify Node.js version: node --version (needs v18+)

4. No Sound Effects

   • Check the Sound Effects toggle in CONFIGURE settings
   • AudioContext may be suspended until first user interaction (click anywhere)

License

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments

• Inspired by 1980s computer systems
• Speech-to-text powered by Groq
• Built with modern web technologies