digestvid 0.2.1

A tool to transcribe and summarize video content.

digestvid

digestvid is a Python package designed to automate the transcription, summarization, and processing of video content. It supports both local video files and YouTube videos, leveraging advanced AI models for accurate transcription and concise summarization. With digestvid, users can efficiently digest video content, making it an ideal tool for content creators, educators, and researchers.

Features

• Transcribe Audio from Videos: Converts spoken words in videos into text using the Whisper model.
• Summarize Transcribed Text: Generates concise summaries from transcribed text using OpenAI's powerful models.
• YouTube Video Support: Downloads YouTube videos for processing, including handling videos with chapters.
• Video Screenshot Capture: Automatically captures and saves a screenshot from the midpoint of a video.
• Flexible Output Management: Saves transcriptions, summaries, and screenshots in user-specified directories, with support for filename sanitization.

Installation

Ensure you have Python 3.6 or later installed. It's recommended to use a virtual environment for installation:

python3 -m venv venv
source venv/bin/activate

Install digestvid using pip:

pip install git+https://github.com/AI-Sherpa/DigestVid.git

Usage

After installation, digestvid can be used from the command line to process either a local video file or a YouTube URL:

digestvid <path to video file or YouTube URL>

Examples

Process a local video file:

digestvid path/to/video.mp4

Download and process a YouTube video:

digestvid https://www.youtube.com/watch?v=dQw4w9WgXcQ

Contributing

Contributions to digestvid are welcome! You can contribute by reporting bugs, requesting features, or submitting pull requests. Please read our contribution guidelines for more information.

License

digestvid is released under the MIT License. See the LICENSE file for more details.