Metadata-Version: 2.4
Name: voicepad
Version: 0.2.1
Summary: Command-line interface for voice recording and GPU-accelerated transcription.
Keywords: cli,voice,audio,transcription,recording,speech-to-text,whisper,command-line,gpu-acceleration
Author: Rajesh Das
Author-email: Rajesh Das <rajesh@hyperoot.dev>
License-Expression: MIT
Classifier: Development Status :: 3 - Alpha
Classifier: Environment :: Console
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: End Users/Desktop
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.13
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Topic :: Multimedia :: Sound/Audio :: Capture/Recording
Classifier: Topic :: Multimedia :: Sound/Audio :: Speech
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Typing :: Typed
Requires-Dist: voicepad-core>=0.2.1
Requires-Dist: typer>=0.23.1
Requires-Dist: rich>=13.0.0
Requires-Dist: textual>=8.2.4
Requires-Dist: pyperclip>=1.9.0
Requires-Dist: keyboard>=0.13.5
Requires-Python: >=3.13
Project-URL: HomePage, https://hyp3r00t.github.io/voicepad/
Project-URL: Repository, https://github.com/HYP3R00T/voicepad
Project-URL: Issues, https://github.com/HYP3R00T/voicepad/issues
Project-URL: Documentation, https://hyp3r00t.github.io/voicepad/
Description-Content-Type: text/markdown

# voicepad

Terminal interface for VoicePad — record audio and transcribe it locally with a single keypress.

## Install

```bash
# Run without installing
uvx voicepad

# Install permanently
uv tool install voicepad
```

## Usage

```bash
voicepad
```

That's it. On first run, the interactive TUI opens an onboarding flow so you can choose a microphone and a starter Whisper model before the first download begins.

| Key | Action |
|---|---|
| `Space` | Start / stop recording |
| `c` | Copy transcription to clipboard |
| `Tab` | Switch tabs (Record / History / Settings) |
| `q` | Quit |

## Configuration

Settings are stored at `~/.config/voicepad/voicepad.yaml` and can be changed from the **Settings** tab inside the app. Some advanced options remain config-only.

```yaml
transcription_model: turbo
transcription_device: auto
transcription_compute_type: auto
input_device_index: null
recordings_path: ~/.config/voicepad/data/recordings
markdown_path: ~/.config/voicepad/data/markdown

# Streaming / chunking
min_chunk_s: 15.0
max_chunk_s: 29.0
overlap_s: 0.5
silence_threshold_ms: 1000
min_fresh_speech_duration_s: 0.25

# Inference cleanup / prompting
initial_prompt: "Hello. This is a transcription with proper punctuation, capitalization, and grammar."
text_postprocessing_enabled: false
no_speech_threshold: 0.6
hallucination_max_repetitions: 3

# VAD tuning
vad_threshold: 0.5
vad_min_speech_duration_ms: 250
vad_speech_pad_ms: 30
```

The UI intentionally exposes a short curated model list for most users. If you want advanced model IDs or lower-level transcription tuning, edit `voicepad.yaml` directly.

## Documentation

**[voicepad.hyperoot.dev](https://voicepad.hyperoot.dev)**
