Get subtitle file for TED video

 

It's a pity that we cant dowload subtitle for videos at TED.COM.  Some people have written 

desktop software and web app to solve the problem.But now these things can't work at all. 

Maybe TED changed  the contents of web page. So I write the script using Perl today.

 It could get subtitle file from TED.COM and convert it to SRT format(SubRip Subtitle File). 

 The script  could just work, I 'll make it perfect next week. Enjoy it!

Last version: https://gist.github.com/949659 

Executable file(Win32): http://www.tinyurl.com/tedsubtitle

Usage: ted.exe URL languageCode output.src 

################################################################################
#  File: ted.pl
#  Desscription: Get ted talk's subtitle from TED.com and convert subtitle of TED video to SRT format(SubRip Subtitle File)
#  Usage: ted.pl URL languageCode output.src 
#  Executable File(Win32): http://www.tinyurl.com/tedsubtitle
#  Create: Thinkhy
#  Date: 2011.04.30   
#  ChangeLog: 1 Add language code. 2011.05.06
#
#  LanguageCode     Language  
#  alb              Albanian
#  ara              Arabic      
#  bul              Bulgarian        
#  chi_hans         Chinese (Simplified)
#  chi_hant         Chinese (Traditional)
#  cze              Czech
#  dut              Dutch
#  eng              English
#  est              Estonian
#  fin              Finnish
#  fre_fr           French (France)
#  ger              German
#  gre              Greek
#  heb              Hebrew
#  hun              Hungarian
#  ita              Italian
#  kor              Korean    
#  pol              Polish   
#  por_br           Portuguese (Brazil)
#  rum              Romanian
#  rus              Russian
#  scr              Croatian
#  spa              Spanish
#  tur              Turkish
#  ukr              Ukrainian
#
################################################################################
##!/usr/local/bin/perl

use strict;
use Data::Dumper;
use JSON;

use LWP::Simple qw(get);

# Magic Number
my $durationOfAdv = 16000; #  seconds of Advertisement time(millisecond).

# Get content from file
# my $content = GetContentFromFile("back.json");

# The TED talk URL
my $url = $ARGV[0];

# languageCode  
my $languageCode = $ARGV[1];

# output file of SRT format
my $outputFile = $ARGV[2];

# !!Note: What you should do is to write URL of TED talks here.
# my $url = "http://www.ted.com/talks/stephen_wolfram_computing_a_theory_of_everything.html";

open OUT, ">out.html";
print "Get html content from URL: $url/n";

# First of all, Get the talkID from the web page.
my $html = GetUrl($url);
$html =~ m/(?<=var talkID = ).*?(/d+)/g;
my $talkID = $1;
chomp($talkID);
print  "TalkID: $talkID/n";

#/(?<=/t)/w+/ 
print OUT $html;
print "Have got html content from URL: $url/n";

# Get subtitle content from TED.COM
my $subtitleUrl = "http://www.ted.com/talks/subtitles/id/$talkID/lang/$languageCode/format/text"; 
my $content = GetUrl($subtitleUrl);

open DEBUG, ">out.json";
print DEBUG $content;

# Decode JSON text
open SRT, ">$outputFile";
my $json = new JSON;
my $obj = $json->decode($content);

my $startTime = $obj->{captions}->[10]->{startTime};
my $duration = $obj->{captions}->[10]->{duration};
my $subtitle = $obj->{captions}->[10]->{content};

my $cnt = 0;
my $len = scalar(@{$obj->{captions}});
print $len;
#foreach my $element ($obj->{captions})
for (;$cnt < $len; $cnt++)
{
    #my %hash = %$element;
    my $startTime = $obj->{captions}->[$cnt]->{startTime};
    my $duration = $obj->{captions}->[$cnt]->{duration};
    my $subtitle = $obj->{captions}->[$cnt]->{content};

    OutputSrt(1+$cnt, $startTime, $duration, $subtitle);
}

###########################################################
# Sub Functions
###########################################################
sub GetTime
{
    my ($time) = @_;

    my $hour = int($time  / 1000 / 3600); 
    my $minute = int((int($time / 1000) % 3600) / 60 ); 
    my $second = int($time / 1000) - $hour * 3600 - $minute * 60; 
    my $msecond = $time - ($hour * 3600 + $minute * 60 + $second) * 1000;

    return ($hour, $minute, $second, $msecond);
}

sub OutputSrt
{
    my ($orderNum, $startTime, $duration, $subtitle) = @_;

    # plus the duration of advertisement
    $startTime += $durationOfAdv;

    # Caculate endTime by duration
    my $endTime = $startTime + $duration; 

    my($hour, $minute, $second, $msecond) = GetTime($startTime); 

    print SRT "$orderNum/n"; # order number

    # Begin time
    print SRT $hour.":".$minute.":".$second.",$msecond";

    # delimitation
    print SRT " --> ";

    # End time
    my($hour1, $minute1, $second1) = GetTime($endTime); 
    print SRT $hour1.":".$minute1.":".$second1.",$msecond/n";

    # Subtitle
    print SRT "$subtitle/n/n";
}


sub GetContentFromFile
{
    my $file = shift; 
    my $content;

    open FILE, $file;  
    while(<FILE>) {
        $content .= "$_";
    }

    return $content;
}

# Test URL: http://www.ted.com/talks/subtitles/id/1130/lang/eng/format/text
sub GetUrl
{
    my $url = shift;
    my $content = get($url);

    return $content;
}