script to get the audio lengths of files in a folder

get_audio_lengths.py

#http://stackoverflow.com/questions/16953842/using-os-walk-to-recursively-traverse-directories-in-python
#http://stackoverflow.com/questions/6239350/how-to-extract-duration-time-from-ffmpeg-output

import os, subprocess, math

#!!!!!!!!!!!!!!! SET THESE TWO PATHS ACCORDING TO YOUR FFMPEG / GAME LOCATION !!!!!!!!!!!!!!!
search_dir=r'C:\games\Steam\steamapps\common\Umineko\voice'
ffprobe_path = r'c:\prog\ffmpeg\ffprobe.exe' #download ffmpeg exe pack from here: https://ffmpeg.zeranoe.com/builds/

file_extensions=[r'.ogg']

def get_partial_path(path, depth):
    total_path = ''

    for i in range(depth):
        rest_of_path, containing_folder = os.path.split(path)
        total_path = os.path.join(containing_folder, total_path)
        path = rest_of_path

    return total_path

FNULL = open(os.devnull, 'w')

for root, dirs, files in os.walk(search_dir):
    file_partial_path = get_partial_path(root, 2) #eg voice/06/

    for file in files:
        _, file_ext = os.path.splitext(file)
        if file_ext not in file_extensions:
            break

        full_file_path = os.path.join(root, file)
        result_as_bytes = subprocess.check_output([ffprobe_path, '-i', full_file_path, '-show_format'], stderr=FNULL)   #stderr=FNULL hides output
        result_as_string = result_as_bytes.decode('utf-8')

        #note: possible to have more than one matching line here, or no matching lines
        matching_lines = [l for l in result_as_string.splitlines() if "duration=" in l]
        tokens = (matching_lines[0]).split('=')
        duration = tokens[1]

        #filename includes partial path, delay is in milliseconds. You may want to add some extra time after each audio plays in your script
        print('{}, {}'.format(os.path.join(file_partial_path,file), math.ceil(float(duration) * 1000)))