v5.1.0 - Major Feature Update

The biggest update to peg_this ever. A complete rewrite of the UI, 20+ new features, AI-powered tools, and a brand new download engine — your studio in a shell just got a serious upgrade.

Highlights

🌐 yt-dlp Download Engine — Download from YouTube, TikTok, Twitter, and 1000+ sites without leaving peg_this
📐 AI Smart Reframe — Auto-crop 16:9 → 9:16 with face tracking for TikTok/Reels
🎙️ AI Auto-Dubbing — Translate and re-voice videos to 24+ languages
🚀 AI Video Upscaling — Real-ESRGAN super-resolution (2x/4x)
🎹 Music Stem Separation — Isolate vocals, drums, bass with Demucs AI
🖥️ Experimental GUI — Full graphical interface with peg_this --gui

Download Engine (yt-dlp)

Download videos, audio, subtitles, and thumbnails from 1000+ sites — all from within peg_this.

5 download modes: Single Video, Audio Only, Playlist, Subtitles Only, Thumbnail Only
Quality + format picker: choose resolution (up to 4K) and container (MP4/MKV/WebM)
Rich video info panel: title, channel (✓ verified), upload date, views/likes/comments, available resolutions, codecs, estimated file size — all shown before you download
Audio extraction: MP3, FLAC, WAV, AAC, Opus, M4A with bitrate selection
Playlist support: download all, first N, range, or specific items with numbering
Advanced options: SponsorBlock, speed limits, time range clips, embed metadata/thumbnail/subtitles/chapters, cookies, proxy
Live progress bar with speed, ETA, and playlist counter
Post-download workflow: immediately trim, compress, convert, extract audio, or generate subtitles on the downloaded file
CLI quick commands:
- peg_this -d <url> — interactive download (pick quality, format)
- peg_this -dy <url> — instant best-quality MP4, zero prompts

AI Features

AI Smart Reframe — Face-tracking auto-crop (16:9 → 9:16, 1:1, 4:5) with temporal smoothing for cinematic panning. Perfect for YouTube → TikTok/Reels/Shorts repurposing
AI Auto-Dubbing — Whisper transcription + translation + Piper TTS voice synthesis (24+ languages)
AI Video Upscaling — Real-ESRGAN models for 2x/4x super-resolution
Music Stem Separation — Demucs AI to isolate vocals, drums, bass, other (2/4/6 stems)
Background Removal — rembg AI for images and videos
Brainrot Captions — TikTok-style word-by-word animated captions
Auto Face Blur — MediaPipe + OpenCV Haar cascade face detection and blurring
Faster-Whisper Subtitles — Multiple model sizes for speed/accuracy tradeoff

Video Features

Trim / Crop / Split / Join — Full editing suite
Aspect Ratio Presets — Quick crop to 16:9, 9:16, 1:1, 4:5, 21:9, and more
Smooth Slow Motion — Optical flow interpolation for buttery slow-mo
Change FPS — Frame rate conversion with smart interpolation
Video Stabilization — Reduce camera shake (vidstab)
Rotation / Flip — 90°/180° rotation, horizontal/vertical flip
Audio Visualizer — Spectrum, waveform, CQT visualization styles
Speed / Reverse / Loop — With RAM estimation and warnings
Color Correction / Denoise / Watermark / PiP / Blur Region
Create GIF — Advanced GIF creation from video

Image Features

Resize / Rotate / Flip / Crop — Full transform suite
Color Adjustment — Brightness, contrast, saturation, gamma
Blur / Sharpen — Gaussian blur and unsharp mask
Effects — Grayscale, sepia, invert
Add Border / Add Text — Custom overlays
Convert / Compress — Format conversion and optimization

Audio Features

Extract / Remove / Merge — Audio manipulation
Volume / Fade / Normalize — Audio adjustment
Format Conversion — Convert between audio formats

Other Features

Experimental GUI — DearPyGui-based graphical interface (--gui flag)
Slideshow Creator — Create video from images with music
Metadata Editor — View and edit video metadata
Batch Convert — Process multiple files at once
File Inspector — Detailed media file analysis
Global Settings — Encoding presets, hardware acceleration, defaults

UI & Architecture

Reorganized Menu — Video → Image → Audio → AI → Other
Dispatch Table Architecture — Clean, maintainable menu routing
Native File Picker — OS-native dialogs
Visual Crop/Blur — OpenCV-based region selection
Keyboard Navigation — Backspace to go back
Centralized Encoding — Global encoder settings with hardware acceleration support

Installation

# Base install
pip install peg_this

# With download support
pip install peg_this[download]

# With AI stem separation (~1.5GB)
pip install peg_this[demucs]

# With AI upscaling (~1.5GB)
pip install peg_this[upscale]

# Everything
pip install peg_this[all-ai,download]

CLI Quick Commands

peg_this                # Interactive menu
peg_this --gui          # Launch GUI
peg_this -d <url>       # Download (interactive)
peg_this -dy <url>      # Download (instant best MP4)

Bug Fixes & Improvements

Fixed unsafe eval() in frame rate parsing — replaced with secure parser
Added RAM estimation and warnings for video reverse operations
Improved aspect ratio crop workflow with presets
Replaced tkinter with OpenCV for visual selection (no more crashes)
Fixed trim seeking accuracy
Improved file handling and resource cleanup

Assets 5

09 Feb 15:24

github-actions

v5.0.1

aadc602

v5.0.1

v5.0.1 - Major Feature Update

This release brings a massive set of new features, a complete UI overhaul, and a brand new experimental GUI.