summarizeme.io - YouTube Transcript Analyzer

This open-source Python Flask application downloads YouTube transcripts (channel-wide) and summarizes them using either:

• OpenAI’s ChatGPT 4o
• Locally Hosted Ollama LLaMA 3.2

All transcripts and summaries are stored in a local folder structure. This project includes asynchronous-like behavior via background threads for demonstration. For production, consider a full task queue solution (Celery, RQ, etc.).

Features

1. Channel-Wide Transcript Download: Provide a channel URL or a video link from the channel. The app iterates over all videos and saves transcripts to data/channels/<channel-id>/transcripts.
2. Separate Summaries:
   • OpenAI => data/channels/<channel-id>/summaries_openai/
   • Ollama => data/channels/<channel-id>/summaries_ollama/
3. Pagination & Filtering: The UI allows filtering by title and basic pagination.
4. Robust Error Handling: Continues downloading/summarizing if a single video fails.
5. Status Updates: The front-end polls status endpoints to show progress (in-progress, completed, failed, etc.).
6. No Database: All data stored on disk.

Folder Structure

my_youtube_transcript_summarizer/
├─ app.py
├─ youtube_utils.py
├─ openai_summarizer.py
├─ ollama_summarizer.py
├─ requirements.txt
├─ .env.example
├─ README.md
├─ data/
│  └─ channels/
│     └─ /
│        ├─ transcripts/
│        │  └─ .json
│        ├─ summaries_openai/
│        │  └─ .md
│        └─ summaries_ollama/
│           └─ .md
├─ templates/
│  ├─ layout.html
│  ├─ index.html
│  ├─ status.html
│  ├─ videos.html
└─ static/
├─ css/
│  └─ styles.css
└─ js/
├─ index.js
├─ status.js
└─ videos.js

Requirements

• Python 3.10+
• Virtualenv or pipenv recommended

Setup

1. Clone the repository:

   git clone [this Repo]
   cd your-repo
   
   

2. Set up a virtual environment:

For Macs:

python3 -m venv venv
source venv/bin/activate

For Windows: For the first time:

python3 -m venv venv
. venv/Scripts/activate

3. Install dependencies:

pip install -r requirements.txt

Create a .env file (based on .env.example):

makefile
Copy code
OPENAI_API_KEY=your-openai-key

4. Run the app:

For Macs:

source venv/bin/activate
flask run

For Windows: For the first time:

. venv/Scripts/activate
flask run

The app will run on http://127.0.0.1:5000.

5. Using the App

Navigate to http://127.0.0.1:5000.

6. (Optional) Ollama: • Make sure Ollama LLaMA 3.2 is installed and running locally if you want to use local summarization.

Usage 1. Open the app in your browser. 2. Enter a YouTube channel URL (or a video from that channel) and click “Start Download.” 3. Check Status to see progress as transcripts are downloaded. 4. Go to the videos page (/videos/) to see the list of videos. 5. Select Summarize (using either OpenAI or Ollama) and track progress. 6. Summaries are saved as .md files in the appropriate folder.