This script convert youtube subtitle file(vtt) to plain text (with no duplicated).

vtt2text

#!/usr/bin/env python3
import re
import os
import glob

def vtt_to_txt(vtt_filepath, txt_filepath):
    """Converts a VTT file to plain text, removing timestamps, header, and duplicates.
    """
    try:
        with open(vtt_filepath, 'r', encoding='utf-8') as vtt_file, \
             open(txt_filepath, 'w', encoding='utf-8') as txt_file:
            previous_line = ""
            for line in vtt_file:
                line = line.strip()
                if line.startswith('WEBVTT') or re.match(r'^[0-9:.]+( --> [0-9:.]+)?$', line) or not line:
                    continue
                if line != previous_line:
                    txt_file.write(line + '\n')
                    previous_line = line
    except FileNotFoundError:
        print(f"Error: File not found: {vtt_filepath}")
    except Exception as e:
        print(f"An error occurred: {e}")

def main():
  vtt_files = glob.glob("*.vtt")
  if not vtt_files:
    print("No .vtt files found in current directory.")
    exit(1)

  for vtt_file in vtt_files:
    txt_file = os.path.splitext(vtt_file)[0] + ".txt"
    if os.path.exists(txt_file):
      print(f"Skipping '{vtt_file}': Output file '{txt_file}' already exists.")
      continue
    vtt_to_txt(vtt_file, txt_file)
    print(f"Converted '{vtt_file}' to '{txt_file}'")


if __name__ == "__main__":
    main()

cp vtt2text ~
sudo ln -s ~/vtt2text /usr/bin
vtt2text /path/to/vttfile

will generate a vttfile.txt within it's relateive path