This repo keeps track of which Destiny VODs are currently transcribed for knowyourinfluencer.io
You can contribute to this project by using Whisperer.py and generating a transcript for a video that has not yet been transcribed. All un-transcribed videos have a deidcated issue. The steps are simple:
- Comment under an issue you wish to claim
- Complete the transcription process described in the How to transcribe section below.
- Submit a pull request with the transcript csv file in the
transcripts/directory, addressing the issue you claimed.
I've made the process as streamlined as I could. All you need is a little programming experience, a GitHub account, and of course the dependencies/hardware requirements needed to run Whisperer.py.
The only way I've provided to generate a transcript is through Docker & docker-compose.
It is possible without Docker, but making it portable is not a simple task, so I've opted to use Docker for the distribution.
You'll need a GitHub account and git installed on your machine for this part.
You'll need to fork the repo so you can later submit a pull request once you've generated a transcript. You can do this by clicking the "Fork" button in the top right corner of the repo page, and cloning it to your machine where you'll be running Whisperer.py.
Once you've claimed an issue, the video_id you'll be generating a transcript for is the title of that issue itself. You'll then use it with docker-compose to generate a transcript.
# template example
VIDEO_ID=<video_id> docker-compose up
# real example
VIDEO_ID=6zK3i3uK-E0 docker-compose upThis will build the Docker image, start the container, and run Whisperer.py with the provided VIDEO_ID. It will output the transcript as transcripts/<video_id>.csv. You can then submit a pull request with the transcript.
Once it's run, it will take quite awile to finish depending on the size of the video, so just leave it running in the background for awhile. When it's finished, commit your changes to your repo with a message resolves #xxx, push it to GitHub, and submit a pull request for the corresponding issue you claimed.
Your pull request should be in the following format:
title: resolves <video_id_issue_number>
description: How you want to be listed as a contributor (e.g. what image, name/username), and any notes you have about the transcript.
For example, the following pull request address issue #256
title: resolves #256
description: Username: EvilFossil, image: https://i.kym-cdn.com/entries/icons/original/000/025/526/gnome.jpg
Whisperer.py also allows you to transcribe Odysee videos. The process is the same as YouTube videos, except instead of passing a YouTube video ID, you pass the entire Odysee URL.
VIDEO_ID=https://odysee.com/@gnomevods:3/RSIejcoWqCU:a docker-compose upIf/when your pull request is approved and merged, I'll update the database to include it, and I'll list you as a contributor at https://knowyourinfluencer.io/contributors/.
Thanks for your help. dggL