Web App to automatic creation of a text publication based on a youtube video
- Backend API on Fast API
- The text from the video is obtained by subtitle extraction paths, or if there are none, by using the Whisper library deployed in a Docker container
- The gpt4free library is used to produce formatted text, and the neural network is used to process raw text
- Class based answer
- Screenshots from video created by OpenCV labrary, may save on server or imgurAPI
- 📌 Сlone repo
- 📎 Install requirements from
requirements.txt
- Docker-composu build \ up for Whisper container