kokseen1 / sanzan Goto Github PK

Possible solutions:

• Pad audio to match video before generating keys
  • Perform comparison between self.mediainfo["duration"] and len(self.audio_raw) to check for mismatch
  • Should have no issue if video is shorter than audio, video will be padded by ffmpeg automatically
• Convert to mp4 before generating keys
• Generate audio key using more reliable methods (e.g. shared key)

TODO:

• Always export keys unless specified otherwise with flag
• Prioritize audio_length from keyfile over audio_length from audio track

Consider multiprocessing