Overview

cr_download checks recent Critical Role Twitch VODs for videos with titles looking like Critical Role episode titles, and prompts the user to download each one. The file is downloaded as video using the streamlink API, and converted to audio using ffmpeg.

Optionally, cr_download can use the Chromaprint music fingerprinting algorithm to attempt to detect the soundtracks Critical Role plays before/after the show, in the opening credits, and during the break. If a good enough set of transition points is found in the episode audio, cr_download will recut the audio to leave out pre/post-show segments, intermission, and (optionally) the announcement section of the episode.

In addition, cr_download can also upload the audio files for the episode to a fixed Google drive folder.

Setup

1. Install ffmpeg if you haven't already.

2. If installing via pip (recommended, if I've actually uploaded the package to PyPI by the time you read this), run pip install cr-download.

   Otherwise, download and extract the repository, and run python setup.py install from the directory you extracted it to.

3. Run streamlink --twitch-oauth-authenticate to authorize streamlink to access your Twitch account. Copy the provided oauth token into your config file, located at ~/.config/cr_download/config.yaml (it will be automatically created the first time you run the program).

Usage

Run critrole_download (it should be added to your path after installation) to display a list of recent Critical Role VODs. You may choose one (or more) to download and convert to audio.

To run the automatic audio editor on the downloaded files, run critrole_download -a.

You can also run autocut_vod FILENAME [FILENAME ...] to run the autocutting tool on a local audio/video file.

To see a full list of options for each command, run critrole_download -h or autocut_vod -h. You can also change the behavior of the program by editing options in the configuration file located at ~/.config/cr_download/config.yaml.