Metadata-Version: 2.1
Name: oswrite
Version: 0.0.2
Summary: CLI tool for processing audio files
Author: Prabin Nepal
License: Apache-2.0
Project-URL: Homepage, https://github.com/nepalprabin/oswrite
Project-URL: Changelog, https://github.com/nepalprabin/oswrite/releases
Project-URL: Issues, https://github.com/nepalprabin/oswrite/issues
Project-URL: CI, https://github.com/nepalprabin/osread/actions
Classifier: License :: OSI Approved :: Apache Software License
Requires-Python: >=3.8
Description-Content-Type: text/markdown
Requires-Dist: click
Requires-Dist: python-docx
Requires-Dist: openai>=1.0
Requires-Dist: simpleaudio; sys_platform == "darwin"
Provides-Extra: test
Requires-Dist: pytest; extra == "test"
Requires-Dist: cogapp; extra == "test"

# oswrite

CLI tool for running audio through the OpenAI whisper API

See oswrite: a CLI tool for running audio through the OpenAI whisper API.

PS: It is inspired by Simon Willamson's [osread](https://github.com/simonw/ospeak)

## Installation

Install this tool using pip:

```python
pip install oswrite
```

## Usage

```
oswrite --audio_file "test.mp3"
```

You will need an OpenAI API key. You can set that as an environment variable:

```
export OPENAI_API_KEY="..."
```

Or you can pass it using --token:

```
oswrite --token "..." --audio_file "test.mp3"
```

## Saving transcribed result

If you want to save transcribed result to a file, additionally you can add <code>output</code> parameter with a filename.

For text file:

```
oswrite --token "..." --audio_file "test.mp3 --output "test.txt"
```

For docx file:

```
oswrite --token "..." --audio_file "test.mp3 --output "test.docx"
```
