项目作者: pablo-var

项目描述 :
Gets the most frequent words in a subtitles srt file before seeing the film/tv show.
高级语言: Python
项目地址: git://github.com/pablo-var/learn-english-words-from-srt.git
创建时间: 2018-01-27T23:46:33Z
项目社区:https://github.com/pablo-var/learn-english-words-from-srt

开源协议:

下载


Function that creates a csv with the words sorted by frequency from a srt file. Only words with more than 1 occurrence.


Stops words not included. Only English.

Test example with the film Artificial Intelligence (Steven Spielberg)

  • test.srt (Subtitles)
  • test.txt (Subtitle text)
  • test.csv (Output)
  • test.ipynb (Notebook with command)

You can run with command prompt like this:

python srt_to_csv.py file_name.srt csv_name.csv

Dependencies:

  • itertools
  • nltk
  • collenctions
  • pandas
  • re, sys

A function in order to learn English efficiently:

  • You select the film/tv show you are going to watch.
  • Apply the srt_to_csv function to the srt file associate.
  • Study the most frequent words you don’t know before watching the film/tv show with the ‘csv_name’ output.
  • Watch the film/tv show and see if you remember the words!
  • It helps for me!

References :