Skip to content

Releases: sebkouba/omarchy-stt

v0.1.1 - Critical Config Fix

06 Jan 11:55
@sebkouba sebkouba
v0.1.1
106e842

Choose a tag to compare

View all tags
v0.1.1 - Critical Config Fix Latest
Latest

Release v0.1.1 - Critical Config Fix

🚨 Critical Bug Fix

This release fixes a critical packaging bug that prevented new AUR users from starting the services. The example configuration file was missing approximately 20 required fields, causing immediate startup failures with TOML parsing errors.

What's Fixed

config.toml.example - Complete Rewrite

The example config now includes all required configuration fields with safe defaults:

Added Missing Sections:

  • ✅ [transcription_corrections] - Acoustic error corrections
  • ✅ [dictation_logging] - Complete logging configuration
  • ✅ [llm] - All conversation history fields (8 previously missing)
  • ✅ [llm.tool_sets] - Tool set definitions
  • ✅ [llm.prompt_tool_mapping] - Prompt-to-tool mappings
  • ✅ [watch] - File watcher for batch transcription
  • ✅ [hotkey] - Complete with modifiers and tap_threshold_ms
  • ✅ [[hotkey.bindings]] - Example hotkey binding arrays

Configuration Philosophy:

  • All fields present with safe defaults
  • Advanced features disabled by default
  • Comprehensive inline documentation
  • Users can start services immediately after only setting microphone device

Before This Release

[llm]
enabled = false
model = "moonshotai/kimi-k2-instruct-0905"
max_tool_calls = 5
# ❌ Missing 8+ required fields - services fail to start

After This Release

[llm]
conversation_history_enabled = false
conversation_history_prompts = ["ask"]
conversation_history_clear_word = "clear"
conversation_history_minutes = 5
conversation_max_turns = 10
conversation_history_dir = "/tmp"
file_chat_enabled = false
file_chat_dir = "~/.config/transcribe-rs/chats"

[llm.tool_sets]
none = []

[llm.prompt_tool_mapping]
# ✅ All required fields present - services start successfully

Impact

Before: New AUR users could not start any services (100% failure rate)
After: Services start successfully with example config after only setting microphone

Testing

This release was thoroughly tested following the complete AUR package installation workflow:

  • ✅ Fresh installation via yay -S omarchy-stt-bin
  • ✅ All three daemons (recording, transcription, hotkey) start successfully
  • ✅ Transcription tested and working with bundled Parakeet model
  • ✅ Configuration validated for all required fields

See lessons-learned/2026-01-06-aur-package-testing.md for the complete test report.

Installation

For New Users

yay -S omarchy-stt-bin

Then follow the post-install instructions to configure your microphone.

For Existing v0.1.0 Users

If you installed v0.1.0 and worked around the config issue:

yay -Syu  # Update to v0.1.1
# Your existing config will continue to work

Files Changed

  • config.example.toml - Complete rewrite with all required fields
  • CHANGELOG.md - New changelog following Keep a Changelog format
  • Cargo.toml - Version bump to 0.1.1
  • packaging/PKGBUILD - Updated for v0.1.1

SHA256 Checksums

4e2c4ffae3c225e81acd3c15671bd2bf16bbd656faf156a220cba6c5fa816676  omarchy-stt-0.1.1-x86_64.tar.gz

Full Changelog

See CHANGELOG.md for complete release history.

This is a critical bug fix release. All v0.1.0 users should upgrade.

Assets 3
Loading

v0.1.0 - Initial AUR Release

06 Jan 10:08
@sebkouba sebkouba
v0.1.0
ca282fa

Choose a tag to compare

View all tags
v0.1.0 - Initial AUR Release

Initial Release

Omarchy STT is a push-to-talk dictation system for Linux/Wayland with hotkey daemon and LLM post-processing.

Features

  • 🎤 Push-to-talk dictation with circular buffer recording
  • ⚡ Fast transcription with Parakeet TDT model (Int8 quantized)
  • 🔑 Global hotkey support via XDG Desktop Portal
  • 🤖 Optional LLM post-processing with Groq (grammar, formatting, tools)
  • 📋 Smart clipboard integration (preserves clipboard content)
  • 🖥️ Terminal detection for correct paste behavior

Installation (Arch Linux)

AUR Package (Recommended)

yay -S omarchy-stt-bin

Follow post-install instructions to configure.

Manual Build

  1. Download omarchy-stt-0.1.0-x86_64.tar.gz below
  2. Extract: tar -xzf omarchy-stt-0.1.0-x86_64.tar.gz
  3. See SETUP.md inside the extracted folder for complete instructions
  4. Or view SETUP.md on GitHub

What's Included

  • 5 binaries: hotkey-daemon, transcribe-daemon, transcribe-client, recording-daemon, list-microphones
  • 3 systemd user services
  • Complete setup documentation (SETUP.md)

System Requirements

  • Linux with Wayland compositor (Hyprland, GNOME, KDE, etc.)
  • FFmpeg, wl-clipboard, ydotool
  • xdg-desktop-portal (for hotkey daemon)

Quick Start

After installation, see SETUP.md for detailed instructions, or:

# 1. Find your microphone
list-microphones

# 2. Copy config
cp /usr/share/omarchy-stt/config.toml.example ~/.config/transcribe-rs/config.toml

# 3. Edit config with your mic device
# 4. Enable services
systemctl --user enable --now recording-daemon transcribe-daemon hotkey-daemon

SHA256 Checksums

3acee3699c46f956f617a2ff90ea7dfefe97cb85d01dd9ebda630100d7bcc731  omarchy-stt-0.1.0-x86_64.tar.gz
Loading