I wrote Video Shuffle Studio in Python with LLM, you can too !

It's time for another Story - this time we look at LLM's harnessed to write code, which can be used to perform complex tasks and save you time with repetitive and mundane actions.

I needed to make some randomised montage videos for a few projects, and there was no viable solution for what i wanted to do - so i used Claude to show how anyone can start writing software with the minimal programming experience. Let LLM's help you to solve problems by writing python ! My SVS toolset can easily create shorts and montages from long form video easily, then you can place it to music or overdub it.

While i am explaining how to use my tool in this video, the principles apply to any software you can imagine, with many useful tips and tricks crammed into this special in depth video guide!

Join this channel to get access to perks:

00:00:00 Introduction
00:02:21 Problem?
00:06:07 Adobe Scripting?
00:08:25 Caught Pretending
00:10:40 Project venv setup
00:13:15 Video Splitting
00:18:39 Simple Shuffle
00:22:32 Size Reward Shuffle
00:31:12 CUDA Acceleration
00:36:16 Resolving Errors/Bugs
00:38:55 Video Splitter (CUDA)
00:40:44 Professional Answers
00:46:38 Shuffle Script
00:50:50 Shuffle Demo
00:53:29 Joiner Script
00:56:04 Harnessing Hallucination
00:57:01 I-Frames
00:59:10 Playing Pretend
01:03:43 Joiner Demo
01:07:34 Tool Instructions

- Workflow Packs:

- SDXL Lora's

- Introducing Playlist (music/video)

- Checkpoint Merging

- cosXL / cosXL-edit conversion

- 3D Generation

- New Diffusion Models (August '24)
Flux1:

Kolors:

SD3-medium:

Stable Cascade:
SDXS-512:
cosXL & cosXL-edit:

- Stable Cascade series:

- Image Model Training

- Music with Audacity

- DJZ custom nodes (aspectsize node)

stable diffusion cascade
stable diffusion lora training
comfyui nodes explained
comfyui video generation
comfyui tutorial 2024
best comfyui workflows
comfyui image to image
comfyui checkpoints
civitai stable diffusion tutorial

An excellent reason to buy an RTX4090 with 24GB of VRAM.

[salomahal]

it took some time but u really grew on me and by now ur probaply my fav creator on ai stuffs thx man

[jakehunter]

LLMs ability to write code is absolutely amazing. I've got no background in the field other than being a lifelong nerd. I got into generative AI and just learning game development from Youtube videos a little over a year ago. Since then with the help of ChatGPT I've started getting into app development in Python. I wrote a simple interface that allowed me to select a folder or a single file and it would append the frame count to the end of each filename using FFMPEG. Very convenient for some of the video workflows for Comfy. I can give ChatGPT my idea and it suggests how we can build it and holds my hand through the hard parts. I've built two different versions of a very functional voice assistant for Windows that works with OpenAIs api. I literally started coding 12 months ago to teach my kid about game development. The worlds changing fast.

[electrolab]

Oh, I see! I once made something similar as a Gradio interface (Didn't share it). Uses ffmpeg to randomly reshuffle a long video into parts of specific length. What I was missing was content awareness (like words spoken context), then I stopped working on it and moved to other things.. some ppl (me) are shifty sometimes 😁- The process of 'arguing' with the AI is funny (and rather lengthy) - But now coding AI is well improved! - Good wrk, everyone!
And as always.

[CORRUPTEDCORE]

Wicked idea mate been following your Comfyui content. Im having a mare installing everything required. Keep getting ' module not found error'. Install-SVS-cuda only worked when i moved it to the python folder, since the install isnt to PATH. Well out my depth, Python newbie bro sorry.