Python script to split an audio file into 29-minute fragments

splitter.py

import datetime
import json
import shlex
import subprocess
import sys


FFMPEG_COMMAND = 'ffmpeg -i "{cap_name}" -ss {start_time} -t 00:29:00 -acodec copy "{cap_name}_part_{number}.{extension}"'
DESIRED_AUDIO_FRAGMENT_LENGTH_SECONDS = 29 * 60 # 29 minutes

# Validate input
if len(sys.argv) != 2:
    print(f"Incorrect number of arguments. The arguments passed are: {sys.argv}")
    sys.exit(-1)

full_filename = sys.argv[1]
filename, extension = full_filename.split(".")


# Find audio duration
ffprobe_cmd = [
    'ffprobe', 
    '-v', 'error', 
    '-show_entries', 'format=duration', 
    '-of', 'json', 
    full_filename
]

print(f"COMMAND: {ffprobe_cmd}")
process = subprocess.run(ffprobe_cmd, stdout=subprocess.PIPE, stderr=subprocess.PIPE, text=True)
output = process.stdout

if output:
    audio_info = json.loads(output)
    duration = float(audio_info['format']['duration'])
    print(f"DURATION: {duration}")
else:
    print("Could not retrieve audio file information")
    sys.exit(1)


# Split audio
number_of_fragments_needed = int(duration // DESIRED_AUDIO_FRAGMENT_LENGTH_SECONDS) + 1
print(f"FRAGMENTS: {number_of_fragments_needed}")
for i in range(number_of_fragments_needed):
    start_time = str(datetime.timedelta(seconds=i * DESIRED_AUDIO_FRAGMENT_LENGTH_SECONDS))
    command_str = FFMPEG_COMMAND.format(cap_name=full_filename, start_time=start_time, number=str(i), extension=extension)
    # Splitting in a format that will work for bash, e.g. without splitting quoted args
    command = shlex.split(command_str)

    print(f"EXECUTING: {command}")
    subprocess.run(command)

print("COMPLETED!")