srt parser and composer

A couple of modules to parse and generate srt files. No external dependencies needed.

• an srt file or srt string (content of an srt file) to json
• an srt file or srt string to plain text
• an srt file or srt string to json with lines and word level timings
• create srt file from json
• parse an srt file into a csv file.

Usage

Install npm install srt_parser_composer

require

var srtParser  = require('srt_parser_composer').parser;
var srtComposer  = require('srt_parser_composer').composer;
var srtToCsv  = require('srt_parser_composer').srtToCsv;

or

var srt  = require('srt_parser_composer');

Parser

Give an srt file it returns json or a plain text of all the lines.

Srt file

See exampel in nroman_door_manual_transcription.srt

1
00:00:00,160 --> 00:00:04,890
There’s this door on the 10th floor I just
hate so much.

2
00:00:04,890 --> 00:00:05,798
Goddammit!

3
00:00:05,799 --> 00:00:11,629
Do you ever get this door wrong? “pretty
regularly.”

4
00:00:11,629 --> 00:00:12,000
How often? “like 30% of the time.”

Srt lines vs file lines

In the code comments I refer both to srt lines and file lines, so I thought I'd clarify.

Srt flines

1
00:00:00,160 --> 00:00:04,890
There’s this door on the 10th floor I just
hate so much.

File lines

While in terms of file lines the following are all considered individual lines.

1

00:00:00,160 --> 00:00:04,890

There’s this door on the 10th floor I just

hate so much.

The srt parser has four possible outputs - srt parser

The parser has 6 main functions,

they both take in an srt file (file path/name) as input parseSrtFileToJson returns a json and parseSrtFileToText returns plain text without timecodes.

You can also parse the content of an srt file, what I refer to as an srt string directly to plain text or json using using parseSrtContentToJson and parseSrtContentToText

parseSrtFileToJson

parseSrtFileToJson returns a json like the one below.

and can be used as follows.

//var parser = require("srt_parser_composer");
var srtParser  = require('srt_parser_composer').parser;
var srtFile = './example/nroman_door_manual_transcription.srt'

srtParser.parseSrtFileToJson(srtFile, function(res){
  console.log(JSON.stringify(res))
});

Example json output

example_output/norman_door_human_transcription.json

  {
    "id": "1",
    "startTime": "00:00:00,160",
    "endTime": "00:00:04,890",
    "text": "There’s this door on the 10th floor I just\nhate so much.\n"
  },
  {
    "id": "3",
    "startTime": "00:00:05,799",
    "endTime": "00:00:11,629",
    "text": "Goddammit!\nDo you ever get this door wrong? “pretty\n"
  },
  {
    "text": "regularly.”\nHow often? “like 30% of the time.”\n",
    "id": "4",
    "startTime": "00:00:11,629",
    "endTime": "00:00:12,000"
  },
  {
    "id": "6",
    "startTime": "00:00:14,290",
    "endTime": "00:00:16,869",
    "text": "Have you seen people misuse it?\nAll the time. Every day. Constantly.\n"
  },
  {
    "id": "8",
    "startTime": "00:00:17,500",
    "endTime": "00:00:18,350",
    "text": "I hate this door.\nMe too Kelsey.\n"
  },
...
]

srt to word accurate time

With a bit of math the parser can also change the granualrity of the timecodes from srt line levels to word level accuracy.

The timecodes are converted from fractional seconds of srt HH:MM:SS,ms (eg 00:00:17,500) to seconds.

Seconds have been chose because that's what's supported by the HTML5 Video API for attribute such as video.currentTime.

The lines of the srt are preserved in the new data structure by generating an array of arrays containing words objects.

such as the following

Example output

Output looks something like this

[
  [
    {
      "start": 4.89,
      "end": 1.88,
      "text": "There’s"
    },
    {
      "start": 4.46,
      "end": 2.74,
      "text": "this"
    },
    {
      "start": 4.029999999999999,
      "end": 2.3099999999999996,
      "text": "door"
    },
    ....
  ]
  ...
]

See ./example_outputsrtJsonToWordLineJson_sample.json for example file.

parseSrtFileToJsonWordsLines

var parser = require('./parser.js');

var srtFile = './example/nroman_door_manual_transcription.srt'
srtParser.parseSrtFileToJsonWordsLines(srtFile, function(res){
  console.log(res)
});

parseSrtContentToJsonWordsLines

var parser = require('./parser.js');
//second use case
var srtStringContent = fs.readFileSync(srtFile).toString();

srtParser.parseSrtContentToJsonWordsLines(srtStringContent, function(res){
  console.log(res)
});

parseSrtFileToText

while parseSrtFileToText returns a plain text like the one below and can be used like so.

var fs = require('fs');
var srtParser  = require('srt_parser_composer').parser;
var srtFile = './example/nroman_door_manual_transcription.srt'

srtParser.parseSrtFileToText(srtFile, function(res){
  console.log(res)
});

Plain text output Example

There’s this door on the 10th floor I just
hate so much.
Goddammit!
Do you ever get this door wrong? “pretty
regularly.”
How often? “like 30% of the time.”
Have you seen people misuse it?
All the time. Every day. Constantly.
I hate this door.
Me too Kelsey.

working with srt string

I was had a use case of a speech to text API that returned an srt file. but rather then returning the file, It was returning a string with the content of the srt file.

So for flexibility, you can also parse the content of an srt file, what I refer to as an srt string directly to plain text or json using.

parseSrtContentToJson and parseSrtContentToText.

And they give the same output as their file opening counterpart described above. Only difference they take in the content of the srt as a string rather then file path/name of the srt file.

Examples

You can find an example srt in the example folder. And some example of the output int he example_output folder.

you can run following command in terminal to try the example

node parser_test.js

Composer

Give an srt json it returns an srt file either as a string content of the srt or as a path to where the srt file as been written to disk.

For example of srt json specs see example_output/norman_door_human_transcription.json

Usage

var fs = require('fs');
var srtComposer  = require('srt_parser_composer').composer;

var srtJson = './example_output/norman_door_human_transcription.json';
var srtJsonContent = JSON.parse(fs.readFileSync(srtJson).toString());

//creating srt string from json
srtComposer.createSrtContent(srtJsonContent, function(srtString){
  console.log(srtString)
})

//creating srt file from json
var JsonToSrtTest = './example_output/composer_test_srt_from_json.srt';

srtComposer.createSrtFile(JsonToSrtTest,srtJsonContent, function(resSrtFilePath){
  console.log(resSrtFilePath)
})

Srt file to CSV

Sometimes you want to share an srt and get some feedback on it. A google spreadsheet is generally the most effective. With this module you can convert your srt to a csv file, so that it can be uploaded to google spreadsheet, or some other use case.

./example_output/demo_test.csv

n	In	Out	Text
1	00:00:00,160	00:00:04,890	There’s this door on the 10th floor I just hate so much.
2	00:00:04,890	00:00:05,798	Goddammit!
3	00:00:05,799	00:00:11,629	Do you ever get this door wrong? “pretty regularly.”
4	00:00:11,629	00:00:12,000	How often? “like 30% of the time.”

Usage

Takes in file path/name of the srt file, the desired name/path for the csv file, and the callback returns the csv file name you specified once it's done writing to disk.

var srtToCsv  = require('srt_parser_composer').srtToCsv;

var demoCsvFile = './example_output/demo_test.csv'

srtToCsv(srtFile, demoCsvFile, function(resCsvPathFileName){
  console.log(resCsvPathFileName);
})

Examples

You can run the example with

node index_test.js