Python parser for transcripts from MS teams

parser.py

import re, sys
from os.path import exists

def parse_file(source):
    parse_file(source, sys.stdout)

def parse_file(input_file, output_file):
    author_index = 1
    text_index = 2
    author = ''
    full_line = ''

    for line in input_file:
        fields = re.split('<|>', line)
        if len(fields) == 5:
            line_author = fields[author_index]
            line_text = fields[text_index]
            if line_author == author:
                full_line += ' ' + line_text
            else:
                output_file.write(full_line + "\n")
                author = line_author
                full_line = line_author[2:] + '\t' + line_text
    output_file.write(full_line + "\n")

def usage():
    sys.exit('''
    usage: python3 parser.py input_file [output_file]

        input_file: the file containing the raw transcript in Web Video Text Tracks (WebVTT) format
        output_file: the destination of the condensed output. If not provided, output will be written to the console.
    ''')

def validate_path(path):
    if not exists(path):
        sys.exit("The specified path %s does not exist or cannot be opened." % (path))

if __name__ == '__main__':
    if len(sys.argv) < 2:
        usage()
    if len(sys.argv) < 3:
        input_path = sys.argv[1]
        validate_path(input_path)
        with open(input_path) as input_file:
            parse_file(input_file, sys.stdout)
    else:
        input_path = sys.argv[1]
        output_path = sys.argv[2]
        validate_path(input_path)
        with open(input_path) as input_file:
            with open(output_path, 'w') as output_file:
                parse_file(input_file, output_file)