/YouTubeChapterGenerator

Make YouTube Chapters from a downloaded Transcript

Primary LanguagePythonMIT LicenseMIT

YouTubeChapterGenerator

Overview

YouTubeChapterGenerator is a tool designed to automate the creation of YouTube chapter markers from video transcripts. It utilizes various Python scripts to clarify, concatenate, and convert transcripts, leveraging OpenAI's GPT-3 for text processing.

Key Components

1. clarify_transcripts.py

This script clarifies YouTube video transcripts using OpenAI's GPT-3 API. It reads transcripts, sends them to GPT-3 for clarification, and saves the processed transcripts.

2. concatenate files.py

Concatenates individual transcript files into a single output file for ease of processing. It adds markers to distinguish between different videos in the output.

3. convert_transcript_to_spr.py

Converts transcript files into a specific format suitable for speech recognition or further processing. This script reads each transcript, processes it using a chatbot function powered by GPT-3, and saves the output in a new format.

Usage

  • Ensure you have Python installed.
  • Install required dependencies: pip install -r requirements.txt (if a requirements file exists).
  • Run each script as needed:
    • python clarify_transcripts.py for transcript clarification.
    • python concatenate files.py for concatenating transcripts.
    • python convert_transcript_to_spr.py for converting transcripts.

Contributions

Contributions are welcome. Please open an issue or submit a pull request for any enhancements.

License

This project is licensed under the MIT License.

Contact

For any queries or assistance, please open an issue in this repository.