christian-byrne/comfy-skills

ASCII Video Production Pipeline

Create ASCII art videos from existing footage or generate original ASCII animations.

Prerequisites

pip install numpy scipy pillow
# Also needs: ffmpeg
# macOS: brew install ffmpeg
# Linux: apt install ffmpeg

Modes

Mode	Description
Video-to-ASCII	Convert existing video to ASCII representation
Audio-reactive	Visuals driven by audio input
Generative	Procedural ASCII scenes and patterns
Hybrid	Mix video conversion with generative overlays
Lyrics/text	Animated text synced to audio
TTS narration	Text-to-speech with visual accompaniment

Pipeline

INPUT → ANALYZE → SCENE_FN → TONEMAP → SHADE → ENCODE

1. INPUT — Video file, audio file, or parameters for generation
2. ANALYZE — Extract frames (video) or audio features (reactive)
3. SCENE_FN — Generate ASCII frame content per frame
4. TONEMAP — Map pixel brightness to ASCII character ramp
5. SHADE — Apply character set and optional ANSI colors
6. ENCODE — Render to video via ffmpeg

ASCII Character Ramps

# Light to dark (standard)
RAMP = " .:-=+*#%@"

# Extended
RAMP_EXT = " .'`^\",:;Il!i><~+_-?][}{1)(|\\/tfjrxnuvczXYUJCLQ0OZmwqpdbkhao*#MW&8%B@$"

# Block characters (highest contrast)
RAMP_BLOCK = " ░▒▓█"

Video to ASCII

import subprocess
import numpy as np
from PIL import Image
import io

def video_to_ascii(video_path, width=120, fps=15):
    """Extract frames and convert to ASCII."""
    cmd = [
        'ffmpeg', '-i', video_path,
        '-vf', f'fps={fps},scale={width}:-1',
        '-f', 'image2pipe', '-vcodec', 'rawvideo', '-pix_fmt', 'gray', '-'
    ]
    # Process frames...
    proc = subprocess.Popen(cmd, stdout=subprocess.PIPE, stderr=subprocess.DEVNULL)
    # Read frame data, convert each pixel to ASCII char from ramp

Render to Video

# Render ASCII frames to video with ffmpeg
# 1. Generate frame images (PIL/Pillow with monospace font)
# 2. Encode:
ffmpeg -framerate 30 -i frames/frame_%04d.png -c:v libx264 -pix_fmt yuv420p output.mp4

# Add audio track:
ffmpeg -i video.mp4 -i audio.mp3 -c:v copy -c:a aac -shortest output_with_audio.mp4

Creative Standards

• First-render excellence — each frame should look good on its own
• Per-section variation — vary character sets, density, and effects across sections
• Cohesive aesthetics — maintain consistent style within a piece
• Frame rate — 15fps for retro feel, 30fps for smooth, 60fps for fluid
• Resolution — 80-120 columns is the sweet spot for readability