Transcription Module:
- Updated the default model to large-v3. Please note that this requires a GPU with approximately 6GB VRAM.
- [BUG] The file counter for transcription didn't work properly; it now counts correctly.
- [BUG] We encountered a UnicodeDecodeError: 'charmap' codec can't decode byte 0x9d at position 61. It is now set to default to UTF-8 when reading the stdout.