This is a web service enabling audio extraction from given intervals in several selected YouTube videos. The program is based on Dash, and the backend is using python3.
Main execution script.
Should be used before running app.py. Input audio file and it cuts it into pieces (default: 20000 ms) to speed up processing time and save memory.
Backend script having the following functions:
- generate and return address of requested audio;
- save quote and generate an record entry to a CSV file, which is displayed on the right side of page (will introduce SQL in future versions).
- Audio processing: reducing noise, channel volume balancing;
- Audio augmentation: combining other audio files (potentially copyright-free music) to generated audio;
- Audio looping;
- Refine input (from one textbox to several);
- Introduce SQL method to integrate all backend data.
- Adding deep learning neural network to automatically add suitable copyright-free music pieces to generated audio.
https://github.com/plotly/dash-object-detection dash_reusable_components.py (in utils) video_engine (folder)