Extract Text From Videos using Python

 Extract Text From Videos using Python

In machine learning, Speech recognition is an interesting task that allows you to recognize the text behind the audio. With the use of voice recognition, we can also extract text from a video. In this article, I will walk you through how to extract text from videos using Python.

Extract Text From Videos using Python

In this section, I will take you through how to extract text from a video using Python. The first step is to download a video. After downloading the videos you need to install two Python libraries:

SpeechRecognition: pip install SpeechRecognition 

moviepy: pip install moviepy

After installing the above two Python libraries you can start with coding. Here is the complete Python program to convert a video into the text:

import speech_recognition as sr 

import moviepy.editor as mp

from moviepy.video.io.ffmpeg_tools import ffmpeg_extract_subclip

num_seconds_video= 52*60

print("The video is {} seconds".format(num_seconds_video))

l=list(range(0,num_seconds_video+1,60))

diz={}

for i in range(len(l)-1):

    ffmpeg_extract_subclip("videorl.mp4", l[i]-2*(l[i]!=0), l[i+1], targetname="chunks/cut{}.mp4".format(i+1))

    clip = mp.VideoFileClip(r"chunks/cut{}.mp4".format(i+1)) 

    clip.audio.write_audiofile(r"converted/converted{}.wav".format(i+1))

    r = sr.Recognizer()

    audio = sr.AudioFile("converted/converted{}.wav".format(i+1))

    with audio as source:

      r.adjust_for_ambient_noise(source)  

      audio_file = r.record(source)

    result = r.recognize_google(audio_file)

    diz['chunk{}'.format(i+1)]=result

After executing the above Python code you need to create a text document to store all the text that has been extracted from the video:

l_chunks=[diz['chunk{}'.format(i+1)] for i in range(len(diz))]

text='\n'.join(l_chunks)

with open('recognized.txt',mode ='w') as file: 

   file.write("Recognized Speech:") 

   file.write("\n") 

   file.write(text) 

   print("Finally ready!")