srt_to_csv.py

# Convert srt file (from eg. pywhispercpp) into csv format with sub start location also in seconds to create Youtube time link.
# Could be useful for annotating livestream recordings from transcription in a spreadsheet.

# Deps (Python 3.11):
# pip install srt

import csv
import srt

def srt_to_csv(srt_file, youtube_link):
    # Open the SRT file
    with open(srt_file, 'r', encoding='utf-8') as file:
        subs = list(srt.parse(file.read()))

    # Create a CSV file and write the headers
    outfile = f"{srt_file}.csv"
    with open(outfile, 'w', newline='', encoding='utf-8') as file:
        writer = csv.writer(file)
        writer.writerow(['subtitle', 'start_time', 'end_time', 'start_s', 'youtube_link'])
        start_time = subs[0].start

        # Iterate over each subtitle and write the data to the CSV file
        for sub in subs:
            seconds_from_start = (sub.start - start_time).seconds
            yt_link = f"{youtube_link}&t={seconds_from_start}s"
            writer.writerow([sub.content, sub.start, sub.end, seconds_from_start, yt_link])

    print(f"CSV file '{outfile}' created successfully.")

srt_to_csv('video1.mp4.srt', 'https://www.youtube.com/watch?v=XXXXXXXX')