Read Subtitle Files as Tabular Data

Read 'SubRip' < https://sourceforge.net/projects/subrip/> subtitle files as data frames for easy text analysis or manipulation. Easily shift numeric timings and export subtitles back into valid 'SubRip' timestamp format to sync subtitles and audio.

srt

The goal of srt is to read SubRip text files as tabular data for easy analysis and manipulation.

Installation

You can install the development version of srt from GitHub with:

# install.packages("remotes")
remotes::install_github("k5cents/srt")

Example

The .srt standard is used to identify the subtitle components for the columns of a data frame:

1. A numeric counter identifying each sequential subtitle
2. The time that the subtitle should appear followed by --> and the time it should disappear
3. Subtitle text itself on one or more lines
4. A blank line containing no text, indicating the end of this subtitle

library(srt)
library(tidyverse)
library(tidytext)
srt <- srt_example()

#> 1
#> 00:01:25,210 --> 00:01:28,004
#> I owe everything to George Bailey.
#> 
#> 2
#> 00:01:28,422 --> 00:01:30,298
#> Help him, dear Father.
#> 
#> 3
#> 00:01:30,674 --> 00:01:33,718
#> Joseph, Jesus and Mary,

These subtitle files are parsed as data frames with separate columns.

(wonderful_life <- read_srt(path = srt, collapse = " "))
#> # A tibble: 2,268 × 4
#>        n start   end subtitle                           
#>    <int> <dbl> <dbl> <chr>                              
#>  1     1  85.2  88.0 I owe everything to George Bailey. 
#>  2     2  88.4  90.3 Help him, dear Father.             
#>  3     3  90.7  93.7 Joseph, Jesus and Mary,            
#>  4     4  93.8  96.4 help my friend Mr. Bailey.         
#>  5     5  96.9  99.5 Help my son George tonight.        
#>  6     6 100.  102.  He never thinks about himself, God.
#>  7     7 102.  104.  That's why he's in trouble.        
#>  8     8 104.  105.  George is a good guy.              
#>  9     9 106.  108.  Give him a break, God.             
#> 10    10 108.  110.  I love him, dear Lord.             
#> # ℹ 2,258 more rows

This makes it easy to perform various text analysis on the subtitles.

wonderful_life %>% 
  unnest_tokens(word, subtitle) %>% 
  count(word, sort = TRUE) %>% 
  anti_join(stop_words)
#> # A tibble: 1,651 × 2
#>    word       n
#>    <chr>  <int>
#>  1 george   216
#>  2 mary      85
#>  3 bailey    74
#>  4 hey       56
#>  5 harry     53
#>  6 yeah      50
#>  7 gonna     45
#>  8 potter    45
#>  9 home      34
#> 10 money     34
#> # ℹ 1,641 more rows

Or uniformly manipulate the numeric time stamps:

wonderful_life <- srt_shift(wonderful_life, seconds = 9.99)

The subtitle data frames can be easily re-written as valid SubRip files.

tmp <- tempfile(fileext = ".srt")
write_srt(wonderful_life, tmp, wrap = FALSE)

#> 1
#> 00:01:35,200 --> 00:01:37,994
#> I owe everything to George Bailey.
#> 
#> 2
#> 00:01:38,412 --> 00:01:40,288
#> Help him, dear Father.
#> 
#> 3
#> 00:01:40,664 --> 00:01:43,708
#> Joseph, Jesus and Mary,