Skip to content

Library for extracting text and timestamps from multiple subtitle files (.ass, .ssa, .srt, .sub, .txt).

License

Notifications You must be signed in to change notification settings

fedecalendino/pysub-parser

Repository files navigation

pysub-parser

Version Quality Gate Status CodeCoverage

Utility to extract the contents of a subtitle file.

Supported types:

For more information: http://write.flossmanuals.net/video-subtitling/file-formats

Usage

The method parse requires the following parameters:

  • path: location of the subtitle file.
  • subtype: one of the supported file types, by default file extension is used.
  • encoding: encoding of the file, utf-8 by default.
  • **kwargs: optional parameters.
    • fps: framerate (only used by sub files), 23.976 by default.
from pysubparser import parser

subtitles = parser.parse('./files/space-jam.srt')

for subtitle in subtitles:
    print(subtitle)

Output:

0 > [BALL BOUNCING]
1 > Michael?
2 > What are you doing out here, son? It's after midnight.
3 > MICHAEL: Couldn't sleep, Pops.

Subtitle Class

Each line of a dialogue is represented with a Subtitle object with the following properties:

  • index: position in the file.
  • start: timestamp of the start of the dialog.
  • end: timestamp of the end of the dialog.
  • text: dialog contents.
for subtitle in subtitles:
    print(f'{subtitle.start} > {subtitle.end}')
    print(subtitle.text)
    print()

Output:

00:00:36.328000 > 00:00:38.329000
[BALL BOUNCING]

00:01:03.814000 > 00:01:05.189000
Michael?

00:01:08.402000 > 00:01:11.404000
What are you doing out here, son? It's after midnight.

00:01:11.572000 > 00:01:13.072000
MICHAEL: Couldn't sleep, Pops.

Cleaners

Currently, 4 cleaners are provided:

  • ascii will translate every unicode character to its ascii equivalent.
  • brackets will remove anything between them (e.g., [BALL BOUNCING])
  • formatting will remove formatting keys like <i> and </i>.
  • lower_case will lower case all text.
from pysubparser.cleaners import ascii, brackets, formatting, lower_case

subtitles = brackets.clean(
    lower_case.clean(
        subtitles
    )
)

for subtitle in subtitles:
    print(subtitle)
0 > 
1 > michael?
2 > what are you doing out here, son? it's after midnight.
3 > michael: couldn't sleep, pops.

Writers

Given any list of Subtitle and a path it will output those subtitles in a srt format.