An electron application that bundles whisper-ai in order to transcribe video files and allow users to generate videos using timestamps from transcribed video.
The project is not finished , i didn't have time to properly build the distribution flow. I use Linux as my daily driver but i decided to build this targeting Windows machines. The application works but there are lots of things to polish to make an nice MVP.
- Router , currently there is no routing system - hash router should be implemented.
- Code architecture , the project could have better folder structure.
- CUDA capabilities , i didn't have time to test it properly so the default transcription device is CPU instead of CUDA.
- No control over output file , currently the exported MP4 file is given random name and its stored on users PC.
- Text view has no real usage.