# Import libraries(pysrt is used for subtitle operations)
import pysrt
import numpy as np

# Store your subtitle file in another object called subs to perform operations with it.
subs = pysrt.open('subsdent.srt')

# Total number of dialogues in my film:
print("Number of dialogues in my short film: ", len(subs))

Number of dialogues in my short film:  356

# To know a particular dialogue at index 'i' :-
i = 10
text = subs[i].text

# Length of any Dialogue at index i
len(subs[i].text)

52

# To know the Start and End Time of a dialougue at index 'i' :-
starttime = subs[i].start
endtime = subs[i].end
# It appears in a format (Hours, Minutes, Seconds, Milliseconds).
# To know any one paramater, use subs[i].start(or)end.hours(or)minutes(or)seconds(or)milliseconds

# To know the duration(in seconds) and length of each dialogue in the subtitle file.
# Here, nod is referred to as 'number of dialogues.'
durations = []
lengths = []
for nod in np.arange(0,len(subs)):
    length = len(subs[nod].text)
    x = subs[nod].start.hours*3600 + subs[nod].start.minutes*60 + subs[nod].start.seconds
    y = subs[nod].end.hours*3600 + subs[nod].end.minutes*60 + subs[nod].end.seconds 
    startduration = x 
    endduration = y 
    duration = endduration - startduration #Seconds
    durations.append(duration) # To add it in the list of durations
    lengths.append(length) # To add it in the list of lengths

# To know the dialogue of maximum length in my subtitle file.
print("The longest dialogue length: ", np.array(lengths).max())
print("This dialogue is:")
print("-----")
for i in np.arange(0, len(subs)):
    if len(subs[i].text) == np.array(lengths).max():
        print(subs[i].text)
        print("-----")

The longest dialogue length:  98
This dialogue is:
-----
<b> I even bought alcohol just because you wanted to drink,</b>
<b> What else do you wanna do?</b>
-----
<b>I said NO to triples that day! But you said Nothing will happen and police won't be there..</b>
-----

# Defined a function that returns me the duration of a dialougue at index 'i'.
def GiveDurationof(i):
    dialogueduration = (subs[i].start.hours*3600 + subs[i].end.minutes*60 + subs[i].end.seconds) - ( subs[i].end.hours*3600 + subs[i].start.minutes*60 + subs[i].start.seconds)
    return dialogueduration

# To know the dialogue with maximum duration in my subtitle file.
for i in np.arange(0, len(subs)):
    if GiveDurationof(i) == np.array(durations).max():
        print(subs[i].text)

<b>(singing a cute telugu song from KOT movie)</b>

# Defining a function to Find the gap from ending of a particular dialogue to the beginning of the next one.
# Using 'i' as argument will mean the gap after 'i'th dialogue ends till 'i+1'th dialougue begins.
def FindGap(i):
    Gap = (subs[i+1].start.minutes*60 + subs[i+1].start.seconds) - (subs[i].end.minutes*60 + subs[i].end.seconds)
    return Gap

# To Store all the gaps in the subtitle file in a list
Gaps = []
for i in np.arange(0, len(subs) - 2):
    Gaps.append(FindGap(i))

# Printing the maximum gap time between two dialougues
print(np.array(Gaps).max())

59

# To Know after what dialogue the maximum gap occurs:-
for i in np.arange(0, len(subs) - 1):
    if FindGap(i) == np.array(Gaps).max():
        print('The maximum gap is occuring after the dialogue {} at index {}: '.format(subs[i].text, i))

The maximum gap is occuring after the dialogue <b>serious.</b> at index 333:

# To change or makes changes to a dialogue at index 'i':-
subs[i].text = 'NEW DIALOGUE!'
# This will change the dialogue to "NEW DIALOGUE!"

subs[i].text

'NEW DIALOGUE!'

# Selecting section of dialogues within a given time interval and shifting all of them 2 seconds prior.
part = subs.slice(starts_after = {'minutes':10, 'seconds':00}, ends_before = {'minutes':19, 'seconds':00})
part.shift(seconds = -2)

from pysrt import SubRipFile

# To selection a section of dialogues that start after a particular minute and ends before another particular minute.
def slice(start, end):
    part = []
    for sub in subs:
        if (sub.start.minutes > start) and (sub.end.minutes < end):
            part.append(sub)
    return SubRipFile(items=part)
# It returns a SubRipFile that contains all the dialogues in the given condition.

# Install 'sarvamai' package
!pip install sarvamai

# Import SarvamAI object from the package
from sarvamai import SarvamAI

# This is the syntax available at Sarvam API Documentation Website! Enter your API code in the specified place.
client = SarvamAI(
	    api_subscription_key="ENTERKEYHERE",
	)

# This is the cake! The "client.text.translate()" function will take 4 arguments.
# The 4 arguments are Input text, Source and Target languages, Speaker Gender. It returns the translated text as output.
response = client.text.translate(
	    input = "",
	    source_language_code="auto",
	    target_language_code="te-IN",
	    speaker_gender="Male"
	)
	
print(response)

# See how the function is called giving all the necessary arguments, coverting all my dialogues into Telugu Language!
translatedtext = []
for i in np.arange(0, len(subs)):
    cleanedsub = subs[i].text
    response = client.text.translate(input=cleanedsub,
                                     source_language_code="auto",
                                     target_language_code="te-IN",
                                     speaker_gender="Male")
    translatedtext.append(response.translated_text)

# After making the necessary changes, Save your subtitle file! There we go!
subs.save('Directory\\Filename.srt')

Using Sarvam Translate to translate my Film subtitles from English to Telugu¶

Getting to know specific details about my Subtitles

Subs is an object that contains set of dictionaries that specifies Text, Start and End timestamps.¶

Replacing/Modifying specific Dialougues

Shifting all (or) few dialogues by some time.

Translating my dialogues into whatever language I want the output to be in(currently Telugu).

References:-¶