Metadata-Version: 2.4
Name: audioreconstructor
Version: 1.1.0
Summary: Command-line audio reconstruction powered by the Audioreconstruction ONNX model
Project-URL: Homepage, https://github.com/rohan-prasen/audioreconstruction
Project-URL: Repository, https://github.com/rohan-prasen/audioreconstruction
Project-URL: Issues, https://github.com/rohan-prasen/audioreconstruction/issues
Author-email: Rohan Prasen Kedari <rohanprasenkedari@gmail.com>
License-Expression: MIT
Keywords: audio,cli,onnx,super-resolution
Classifier: Development Status :: 4 - Beta
Classifier: Environment :: Console
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: Microsoft :: Windows
Classifier: Operating System :: POSIX :: Linux
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.10
Classifier: Programming Language :: Python :: 3.11
Classifier: Programming Language :: Python :: 3.12
Classifier: Programming Language :: Python :: 3.13
Classifier: Topic :: Multimedia :: Sound/Audio
Requires-Python: >=3.10
Requires-Dist: click>=8.1
Requires-Dist: rich>=13.7
Description-Content-Type: text/markdown

# audioreconstructor

`audioreconstructor` is the command-line release of Audioreconstruction's ONNX
audio enhancer. It installs a small Python launcher; the ONNX model and the native
runtime are downloaded only when you explicitly run setup.

## Install

```bash
pip install audioreconstructor
audioreconstructor setup
```

`setup` downloads the native executable, `model.onnx`, and `config.json` from the
GitHub Release matching the installed package version. It verifies SHA-256 hashes
before making them available locally.

The current release supports 64-bit Linux and 64-bit Windows.

## Use

Enhance a single file:

```bash
audioreconstructor enhance --input song.mp3 --output song_enhanced.flac
```

Enhance a whole folder (recurses into subfolders):

```bash
audioreconstructor enhance --folder ./songs
```

Reconstructed files are written to a mirrored tree under `<folder>/enhanced/`, e.g.
`songs/rock/track.mp3` → `songs/enhanced/rock/track.flac`.

Choose the ONNX execution provider when needed:

```bash
audioreconstructor enhance --input song.mp3 --output song_enhanced.flac --provider cpu
```

`auto` is the default. Windows first tries DirectML and falls back to CPU; Linux uses
CPU. The native executable reads supported audio through libsndfile and always writes
FLAC output.

> **Note:** v1.0.0 used top-level flags (`audioreconstructor --setup`,
> `audioreconstructor --input ... --output ...`). These are now `setup`, `doctor`, and
> `enhance` subcommands. Run `audioreconstructor --help` for the full command list.

## Verify installation

```bash
audioreconstructor doctor
```

Doctor verifies all cached release assets and runs an actual synthetic-audio ONNX
inference test. A healthy installation exits with status `0`.

## Cache locations

Setup prints the exact locations it uses. By default they are:

- Linux: `$XDG_CACHE_HOME/audioreconstructor/<version>` or
  `~/.cache/audioreconstructor/<version>`
- Windows: `%LOCALAPPDATA%\\audioreconstructor\\Cache\\<version>`

Running setup for an upgraded package version removes older cached versions after the
new version has been fully verified.
