mfm-transcripts

This repo uses Python to extract YouTube channel video URLs and mass transcript data from those videos.

Download or clone this repo to get started git clone https://github.com/tpcav/mfm-transcripts.git. Or see the main.py file which has the code for the YouTube Transcript API for Python.

I. Packages

• JSON Decoder import json
• YouTube Transcript API pip install youtube-transcript-api
• Once downloaded to start DEMO run: main.py

II. Getting all channel video URLs without YouTube API

1. Go to the YouTube Channel's video page
2. Inspect Element
3. Go to the console
4. Copy this first var scroll = setInterval(function(){ window.scrollBy(0, 1000)}, 1000); This code scrolls to the bottom of the channels video page, to the first video.
5. Copy this seconde window.clearInterval(scroll); console.clear(); urls = $$('a'); urls.forEach(function(v,i,a){if (v.id=="video-title-link"){console.log('\t'+v.title+'\t'+v.href+'\t')}}); This code returns a list in the console of all of the titles and URLs of all the videos of a channel.

III. Format the links Google Sheets, Excel, etc

For example:

1. This is a full URL https://www.youtube.com/watch?v=2dOCPr355TQ
2. We just want the last part 2dOCPr355TQ This is the video id.
3. The video ids can be used by the YouTube Transcript API to get the transcript data

IV. Using youtube-transcript-api

See main.py

VI. Video Tutorial

TODO, add video file