🎬 YouTube Shorts Creator — Setup & Usage Guide

What It Does

Downloads viral clips → cuts funniest moments → crops to 9:16 vertical → adds meme-style captions → exports ready-to-upload YouTube Shorts MP4s. Bonus: auto-builds a Top-5 compilation video from all clips.

---

Step 1 — Install System Tools

macOS (Homebrew)

brew install ffmpeg yt-dlp

Ubuntu / Debian

sudo apt update
sudo apt install ffmpeg -y
sudo curl -L https://github.com/yt-dlp/yt-dlp/releases/latest/download/yt-dlp \
     -o /usr/local/bin/yt-dlp
sudo chmod a+rx /usr/local/bin/yt-dlp

Windows (PowerShell, as Admin)

# Install Scoop first: https://scoop.sh
scoop install ffmpeg yt-dlp

Verify:

ffmpeg -version
yt-dlp --version

---

Step 2 — Python Environment

# Create venv (recommended)
python3 -m venv venv
source venv/bin/activate      # macOS/Linux
# venv\Scripts\activate       # Windows

# Install Python packages
pip install moviepy pillow numpy yt-dlp

Optional — AI auto-captions (Whisper):

pip install openai-whisper

(Whisper integration can be added by calling whisper <clip.mp4> and piping the SRT into ffmpeg's subtitle filter.)

---

Step 3 — Project Layout

After running the script once, this structure is created automatically:

project/
├── shorts_creator.py   ← the main script
├── downloads/          ← raw downloaded videos
├── clips/              ← intermediate cut/vertical clips
├── output/             ← ✅ FINAL YouTube Shorts go here
│   ├── video1.mp4
│   ├── video2.mp4
│   └── TOP5_compilation.mp4
└── fonts/              ← Roboto-Bold.ttf (auto-downloaded)

---

Step 4 — Run It

Default (uses the 5 hardcoded examples)

python3 shorts_creator.py

Add a custom clip via CLI

python3 shorts_creator.py \
    "https://www.youtube.com/watch?v=YOUR_VIDEO_ID" \
    12.5 \
    27.0 \
    "WAIT FOR IT 😂" \
    my_funny_clip

Arguments: <url> <start_sec> <end_sec> <caption_text> <output_name>

---

Step 5 — Customize the Clip List

Edit the CLIPS_LIST section in shorts_creator.py:

CLIPS_LIST: list[ClipSpec] = [
    ClipSpec(
        url     = "https://www.youtube.com/watch?v=YOUR_VIDEO",
        start   = 10.0,    # start time in seconds
        end     = 23.0,    # end time in seconds (5–15 sec clips only)
        caption = "THIS IS INSANE 😂🔥",
        title   = "my_clip_name",  # used for filenames
        zoom    = 1.15,    # subtle zoom-in (1.0 = none, 1.3 = big)
    ),
    # add more ClipSpec(...) entries here
]

---

Step 6 — Download Your Shorts

After the script finishes, collect everything from the output/ folder:

ls -lh output/
# video1.mp4    → Short #1
# video2.mp4    → Short #2
# ...
# TOP5_compilation.mp4  → Bonus compilation

Upload directly to YouTube → Create → Upload Video → select Shorts format.

---

Configuration Reference

Variable	Default	Description
SHORTS_W / H	1080×1920	YouTube Shorts resolution
CAPTION_FONT_SIZE	72	Caption text size in pixels
CAPTION_STROKE_W	6	Black outline thickness
CAPTION_POSITION	0.72	Vertical position (0=top, 1=bottom)
TARGET_FPS	30	Output frame rate
CRF	23	Video quality (18=best, 28=smallest)
PRESET	medium	ffmpeg speed/quality tradeoff

---

Troubleshooting

Problem	Fix
yt-dlp rate-limited / blocked	Add --cookies-from-browser chrome to yt-dlp cmd in script
zoompan makes video slow	Reduce clip zoom to 1.05 or 1.0
Captions look bad	Increase CAPTION_FONT_SIZE or change CAPTION_POSITION
Audio out of sync	Change CRF from 23 → 18 for higher quality
moviepy ImportError	pip install moviepy==1.0.3 (stable version)
Font not loading	Place Roboto-Bold.ttf manually in fonts/ folder

---

Example URLs (Safe Public Domain / Well-Known)

# Rick Astley - Never Gonna Give You Up
https://www.youtube.com/watch?v=dQw4w9WgXcQ

# Me at the zoo (first YouTube video)
https://www.youtube.com/watch?v=jNQXAC9IVRw

# Charlie Bit My Finger
https://www.youtube.com/watch?v=FtutLA63Cp8

# Evolution of Dance
https://www.youtube.com/watch?v=e_DqV1xdf-Y

⚠️ Copyright notice: Only use clips you have rights to, or clips under Creative Commons license. Use yt-dlp --list-formats <url> to inspect available quality options.

---

Optional: Auto-Captions with Whisper

pip install openai-whisper

# Generate SRT for a clip:
whisper clips/my_clip_raw.mp4 --model base --output_format srt

# Then burn into video with ffmpeg:
ffmpeg -i clips/my_clip_vertical.mp4 \
       -vf "subtitles=clips/my_clip_raw.srt:force_style='FontSize=24,PrimaryColour=&Hffffff&'" \
       output/video1.mp4