Morse Bot

A real-time morse code decoder and encoder built as a TypeScript monorepo. Decode morse code from your microphone or audio files, view live signal visualizations, and encode text back to morse with audio playback.

Features

• Real-time microphone decoding - Listen to morse code through your mic and see it decoded live
• Audio file decoding - Upload WAV files and decode morse code with progress tracking
• Adaptive speed tracking - Automatically adapts to the sender's WPM with EMA smoothing
• Live spectrogram - Canvas-based scrolling waterfall display of the audio signal
• Text-to-morse encoder - Convert text to morse code with audio playback and character highlighting
• Session persistence - Save, view, and export decoded sessions (authenticated users)
• Configurable parameters - Tune frequency (300-1200 Hz), WPM, detection threshold, and adaptive mode

Architecture

This is a Turborepo monorepo using the @morse-bot package scope.

apps
  ├─ nextjs          — Next.js web app with decoder UI, auth, and session management
  ├─ expo            — React Native mobile app (Expo SDK 54)
  └─ tanstack-start  — Tanstack Start web app
packages
  ├─ morse-decoder   — Core decoding/encoding library (Goertzel filter, envelope detector,
  │                    timing analyzer, morse tree, encoder)
  ├─ api             — tRPC v11 router (posts, sessions)
  ├─ auth            — Authentication via Better Auth
  ├─ db              — Drizzle ORM with Supabase/Postgres
  ├─ ui              — Shared UI components (shadcn/ui)
  └─ validators      — Shared validation schemas
tooling
  ├─ eslint          — Shared ESLint presets
  ├─ prettier        — Shared Prettier config
  ├─ tailwind        — Shared Tailwind theme
  └─ typescript      — Shared tsconfig

Morse Decoder Pipeline

Audio samples flow through a signal processing pipeline:

1. Goertzel filter - Single-frequency IIR detector (no FFT needed)
2. Envelope detector - Dual-threshold hysteresis tone classifier with adaptive noise floor
3. Timing analyzer - Classifies durations into dits, dahs, and gaps (element/character/word)
4. Morse tree lookup - Binary tree mapping of dit/dah sequences to ITU characters

Quick Start

Prerequisites

• Node.js ^24.13.1
• pnpm ^10.19.0

Setup

# Install dependencies
pnpm i

# Configure environment variables
cp .env.example .env

# Push the database schema
pnpm db:push

# Generate the Better Auth schema
pnpm auth:generate

# Start development
pnpm dev

Running Tests

pnpm test

Tests validate the morse decoder against sample WAV files with an 80% accuracy target. Current results:

File	Expected	Accuracy
morse-test1.wav	"HI. HOW ARE YOU?"	90.3%
morse-test2.wav	"THE ORANGE FOX JUMPS OVER THE BROWN LOG."	98.7%
morse-test3.wav	"WHERE IS THE BACON?"	97.3%

Deployment

The Next.js app can be deployed to Vercel. Select apps/nextjs as the root directory and add your POSTGRES_URL environment variable.

License

MIT