Simple karaoke-style video generator, using whisper for transcription and whisperx for token-wise text alignment.
-
GPU-equipped machine
-
ffmpeg
-
pip install -r requirements.txt
$ python karaokit.py media_file{mp3 or mp4} output_file_dir
karaokit.karaokit
: Generate a subbed video from an media file.
- ja_song.mp3 from BGMusic