Download YouTube videos as MP3s from URLs

download_youtube_audio.py

# You will need to:
#   pip install yt-dlp
# Put YouTube URLs in a text file, 
# optionally followed by a tab and a number for destination directories, for example:
# https://www.youtube.com/watch?v=l7A0lDnlRMQ	15
# https://www.youtube.com/watch?v=x1VMJKOz9SI	15
# Usage:
#   python download_youtube_audio.py


import os
import re
from yt_dlp import YoutubeDL

# Path to your text file containing the YouTube URLs and numbers
input_file = 'youtube_urls.txt'
# Path to the file for logging failed downloads
failed_log_file = 'failed_downloads.txt'

# Base directory where MP3 files will be saved
base_output_dir = 'downloaded_mp3s'
unsorted_dir = os.path.join(base_output_dir, 'unsorted')
os.makedirs(base_output_dir, exist_ok=True)
os.makedirs(unsorted_dir, exist_ok=True)

# Characters to be removed
remove_chars = r"[ ,;:'\"`~!@#$%^&*()\[\]{}?/\\|=+<>-]"

def sanitize_filename(title):
    # Replace spaces and hyphens with underscores
    title = re.sub(r"[ -]", "_", title)
    # Remove punctuation and non-alphanumeric characters
    title = re.sub(remove_chars, "", title)
    return title

def download_audio(url, output_dir):
    ydl_opts = {
        'format': 'bestaudio/best',
        'outtmpl': os.path.join(output_dir, '%(title)s.%(ext)s'),
        'postprocessors': [{
            'key': 'FFmpegExtractAudio',
            'preferredcodec': 'mp3',
            'preferredquality': '192',
        }],
    }

    try:
        with YoutubeDL(ydl_opts) as ydl:
            info_dict = ydl.extract_info(url, download=False)
            title = sanitize_filename(info_dict['title'])
            ydl_opts['outtmpl'] = os.path.join(output_dir, f'{title}.%(ext)s')
            with YoutubeDL(ydl_opts) as ydl:
                ydl.download([url])
        print(f'Successfully downloaded and converted: {url}')
    except Exception as e:
        print(f'Error downloading {url}: {e}')
        with open(failed_log_file, 'a') as log_file:
            log_file.write(url + '\n')

with open(input_file, 'r') as file:
    lines = file.readlines()

for line in lines:
    # Skip blank lines
    if not line.strip():
        continue

    parts = line.strip().split('\t')
    url = parts[0].strip()
    number = parts[1].strip() if len(parts) > 1 else None

    if url:
        if number:
            output_dir = os.path.join(base_output_dir, number)
        else:
            output_dir = unsorted_dir
        os.makedirs(output_dir, exist_ok=True)
        download_audio(url, output_dir)
    else:
        print(f'Invalid line: {line.strip()}')