https://github.com/rafaykhattak/youtubegpt

YoutubeGPT is a web application powered by OpenAI's Whisper model for speech recognition and GPT-3 for text summarization. It extracts transcriptions from YouTube videos and provides users with concise summaries based on the video content.
https://github.com/rafaykhattak/youtubegpt

deep-neural-networks ffmpeg gpt-3 pytube whisper-ai

Last synced: 3 months ago
JSON representation

YoutubeGPT is a web application powered by OpenAI's Whisper model for speech recognition and GPT-3 for text summarization. It extracts transcriptions from YouTube videos and provides users with concise summaries based on the video content.

• Host: GitHub
• URL: https://github.com/rafaykhattak/youtubegpt
• Owner: RafayKhattak
• Created: 2023-05-17T09:04:29.000Z (over 1 year ago)
• Default Branch: main
• Last Pushed: 2023-05-17T09:31:29.000Z (over 1 year ago)
• Last Synced: 2024-08-05T06:04:47.454Z (6 months ago)
• Topics: deep-neural-networks, ffmpeg, gpt-3, pytube, whisper-ai
• Language: Python
• Homepage: https://rafaykhattak-youtubegpt-app-1ovyiq.streamlit.app/
• Size: 6.84 KB
• Stars: 17
• Watchers: 2
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# YoutubeGPT

YoutubeGPT is a web application that utilizes OpenAI's state-of-the-art models to enhance the accessibility and comprehension of YouTube videos. By leveraging the power of Whisper, a general-purpose speech recognition model, and GPT-3, a language model capable of generating human-like text, YoutubeGPT extracts transcriptions from YouTube videos and provides users with concise and accurate summaries based on the video content.

![Screenshot (455)](https://github.com/RafayKhattak/YoutubeGPT/assets/90026724/7eb66cf6-73db-490c-9212-2e2ffd76bae7)

## Features

- **Speech Recognition**: YoutubeGPT employs OpenAI's Whisper model to transcribe the audio content of YouTube videos, ensuring accurate and reliable speech-to-text conversion.

- **Summarization**: Leveraging GPT-3, YoutubeGPT generates concise summaries of the video content, enabling users to quickly grasp the main points without having to watch the entire video.

- **User-Friendly Interface**: The web application offers a user-friendly interface where users can simply enter the URL of the YouTube video they want to process and receive the transcription and summarized text promptly.

## Getting Started

To run YoutubeGPT locally, follow these steps:

1. Clone this repository to your local machine.

2. Install the required dependencies by running `pip install -r requirements.txt`.

3. Install ffmpeg, which is required for audio processing. You can download ffmpeg from the official website (https://ffmpeg.org/) and add it to the environment variable path.

4. Set up your OpenAI API key by following the instructions in the `.env.example` file.

5. Run the application using the command `streamlit run app.py`.

6. Access the web application in your browser at `http://localhost:8501`.

## Technologies Used

- Python

- OpenAI Whisper

- OpenAI GPT-3

- Streamlit

## Limitations

Please note the following limitations of YoutubeGPT:

- The accuracy of speech recognition may vary depending on the quality of audio in the YouTube video.

- GPT-3 summarization is based on the extracted transcription and may not capture all nuances or context from the video.

- Due to OpenAI API limits, there may be restrictions on the number of API calls that can be made.

## License

YoutubeGPT is licensed under the [MIT License](https://opensource.org/licenses/MIT). Feel free to use, modify, and distribute the code as per the terms of the license.

## Acknowledgements

- This project utilizes OpenAI's Whisper and GPT-3 models. Special thanks to OpenAI for providing access to these powerful AI technologies.

- The web application is built using Streamlit, a fantastic framework for creating interactive data applications in Python.