Metadata-Version: 2.4
Name: whisper-assistant
Version: 0.1.10
Summary: Voice transcription tool using Groq API with global hotkeys
License-File: LICENSE
Requires-Python: >=3.10
Requires-Dist: click>=8.1.7
Requires-Dist: groq>=0.9.0
Requires-Dist: numpy>=1.26.4
Requires-Dist: pyaudio>=0.2.14
Requires-Dist: pynput>=1.7.6
Requires-Dist: pyperclip>=1.8.2
Requires-Dist: python-dotenv>=1.0.1
Requires-Dist: soundfile>=0.12.1
Description-Content-Type: text/markdown

# Whisper Assistant

Voice-to-text tool powered by Groq's Whisper API. Press a hotkey, speak, and get instant transcription.

## Get Started

Requires macOS and [uv](https://github.com/astral-sh/uv).

```bash
# Install portaudio (required for audio recording)
brew install portaudio

# Run the setup wizard
uvx whisper-assistant init
```

The wizard will:
1. Ask for your [Groq API key](https://console.groq.com/keys) (free)
2. Configure hotkeys and preferences
3. Start the background daemon

**Where files are stored** ([XDG Base Directory Specification](https://xdgbasedirectoryspecification.com/)):
| Path | Contents |
|------|----------|
| `~/.config/whisper-assistant/config.env` | API key, hotkeys, preferences |
| `~/.local/share/whisper-assistant/history/` | Audio recordings & transcriptions |
| `~/.local/state/whisper-assistant/logs/` | Log files |
| `~/.local/state/whisper-assistant/daemon.pid` | Daemon process ID |

## Usage

```bash
# Daemon control
whisper-assistant start
whisper-assistant stop
whisper-assistant restart
whisper-assistant status

# View logs
whisper-assistant logs
whisper-assistant logs --stderr

# Configuration
whisper-assistant config show
whisper-assistant config edit

# History
whisper-assistant history list
whisper-assistant history play 1      # play most recent recording
whisper-assistant history transcribe 1  # re-transcribe most recent

# Transcribe any audio file
whisper-assistant transcribe /path/to/audio.wav
```

**Default hotkeys:**
- `Ctrl+Shift+1` — Toggle recording (press to start, press again to stop and transcribe)
- `Ctrl+Shift+2` — Retry last transcription
- `Ctrl+Shift+3` — Cancel recording

**Config location:** `~/.config/whisper-assistant/config.env`

## Contributing

PRs welcome. For major changes, open an issue first.

```bash
git clone https://github.com/Futhark-AS/whisper_assistant.git
cd whisper_assistant
uv sync
uv run whisper-assistant --help
```

[MIT License](LICENSE)
