db - downbad

A minimal markdown editor for macOS built with Rust and egui.

db <filename>

Opens the file in a raw text editor with line numbers. The process detaches from the terminal so your shell prompt returns immediately.

Editing	Preview (Cmd+P)

Installation

cargo install --path .

This builds a release binary and installs db to ~/.cargo/bin/. Make sure ~/.cargo/bin is in your PATH.

Keyboard Shortcuts

Key	Action
Cmd+S	Save
Cmd+E	Exit (prompts if unsaved)
Cmd+P	Toggle markdown preview
Cmd+R	Reload file from disk
Cmd+D	Toggle speech-to-text recording
Cmd+T	Text-to-speech (selection or document)
Cmd+C/V/X	Copy / Paste / Cut

Unsaved Changes Dialog

Key	Action
S	Save & Exit
D	Discard & Exit
Esc	Cancel

Speech-to-Text (Whisper)

Cmd+D starts recording from the default microphone. Press Cmd+D again to stop and transcribe. The Whisper model is loaded lazily on first use, so it never slows down app startup.

Setup (one-time)

mkdir -p ~/.local/share/downbad
curl -L -o ~/.local/share/downbad/ggml-base.en.bin \
  https://huggingface.co/ggerganov/whisper.cpp/resolve/main/ggml-base.en.bin

Text-to-Speech (Kokoro)

Cmd+T speaks the selected text (or full document if nothing is selected) aloud using the Kokoro-82M TTS model running locally via ONNX Runtime. Press Cmd+T again to stop playback.

Setup (one-time)

Download the three model files into ~/.local/share/downbad/:

mkdir -p ~/.local/share/downbad
# ONNX model (quantized, ~92 MB)
curl -L -o ~/.local/share/downbad/kokoro-v1.0.onnx \
  https://huggingface.co/onnx-community/Kokoro-82M-v1.0-ONNX/resolve/main/onnx/model_quantized.onnx
# Tokenizer
curl -L -o ~/.local/share/downbad/kokoro-tokenizer.json \
  https://huggingface.co/onnx-community/Kokoro-82M-v1.0-ONNX/resolve/main/tokenizer.json
# Voice style (af_heart)
curl -L -o ~/.local/share/downbad/kokoro-voice.bin \
  https://huggingface.co/onnx-community/Kokoro-82M-v1.0-ONNX/resolve/main/voices/af_heart.bin