Change in readme.md

[ArqiesAr]

gi to git

[what-the-diff]

• Renamed the project to auto-subtitle-plus
• Added support for multiple languages (see --language parameter)
• Fixed audio out of sync issue by using ffmpeg instead of pydub and audiosegment
• Wildcard support for filenames, so you can use it like this: auto_subtitle *.mp4 or even with a folder path as input: auto_subtitle /path/to/folder/*.*
• Convert audio to subtitles (output .srt files), useful if you want to generate srt's without generating videos at all, just run something like this : auto_subtitles video1*.avi -o outputdir --output-srt. This will create an .srt file next to your original video in the specified directory (-o). You can also specify only one filename too!
• Add a new function to check if the input file is an audio
• Add a new function to extract audio from video using ffmpeg
• Change the main() method in cli.py:
  • Extract audios first, then generate subtitles with whisper and finally add them into videos (if output_video=True)
• Added a new file utils.py
• Updated requirements to include psutil and youtube-dl
• Changed the name of the package from auto_subtitle to auto_subtitle_plus, updated author information, added description for console script entry point in setup.py

[thebetauser]

Hello, can you please add a check for using Apple silicon (M1/M2/M3)

if torch.cuda.is_available():
    default_device = "cuda"
elif hasattr(torch.backends, 'mps'):
    if torch.backends.mps.is_available():
        default_device = "mps"
else:
    default_device = "cpu"

parser.add_argument("--device", default="cuda" if torch.cuda.is_available() else ("mps" if hasattr(torch.backends, 'mps') and torch.backends.mps.is_available() else "cpu"), help="device to use for PyTorch inference")

Edit:
Whisper is currently adding support for this so as of 7/14 it will not work.