Video-Notes is a SaaS tool that automatically generates notes from informational videos. It utilizes advanced technologies such as automatic transcription, language translation, and natural language processing to provide users with concise summaries, transcripts, and even blog posts based on the video content.
- Automatic transcription of video content
- Language translation for transcripts
- Summarization of video content
- Generation of blog posts from video content
Video-Notes is built using the following technologies and frameworks:
- Authentication: Passage
- Fetching audio from YouTube: YTDL
- Transcription: YouTube's auto-cc (previously Whisper API)
- Language model: ChatGPT (GPT-3.5 architecture)
- Data storage: Firebase
To run Video-Notes locally, follow these steps:
- Clone the repository:
git clone https://github.com/theashishmaurya/videoNotes.git - Install the required dependencies:
yarn - Set up authentication using Passage. Refer to the documentation for more details.
- Configure Firebase for data storage. Refer to the documentation for detailed instructions.
Contributions are welcome! If you have any ideas, bug fixes, or improvements, please submit a pull request. Make sure to follow the repository's code of conduct.
Video-Notes is open source and released under the MIT License.
Check out the live version of Video-Notes at https://video-notes-pi.vercel.app.
For detailed usage instructions and examples, please refer to the Blog.
If you have any questions, suggestions, or feedback, feel free to contact the development team at [email protected].