Volumetric video! Nice!!

Imagine a next-gen VR experience that lets you speak realistic scenes, intelligent characters and complex situations into being, then interact with them in real time. It’s coming, due to a convergence of tech like this advance in real-time 3D video.

https://newatlas.com/vr/holodeck-real-time-volumetric-3d-video/

https://x.com/minchoi/status/1868137176254570810

Some videos in here that show the 3d point clouds that are generated.

I think in this blog for CWT I mention a possibility like this - viewers having the ability to transport into another character’s perspective, change setting, culture, etc.

I think I wrote about this. If I didn’t I am now. Part of the CWT is to enhance content with the writer or creator’s surroundings. It would be cool to be able to zoom away from the content itself and put yourself in the creator’s shoes. Where were they when they sketched the idea. Or wrote the thing. Or made the music. Etc. This does that:

https://x.com/BradSmi/status/1834991383964533237

Old draft post below posted for - why not

Creative Enhancement Engine

Premise

An engine that lets you send your content to it, send your thoughts about what to do with the content, and have it recommend some prompts or choices based on the content type? Some fine tuning would be good for this. Some analysis of content types specific to Spoke would make it develop posts similar to other Spoke’s style. However, it could choose from a list of prompts like SK does.

Challenge

So then a challenge is to build up the prompts. Build up the skills. And let your program access those in a simple way.

Promise

Embedding this service into Spoke or another application. Return ideas or options to user on the fly. Could spawn a worker AI container/spark/etc. to make it happen.

At some level, I want a simple AI machine that I can whisper things to and let it store that for me. So that later on I can possibly retrieve the info in some connected way.

Generate Imagery/Video/Audio all at once, in one model… CWT? 😢

https://codi-gen.github.io/static/images/teaser.mp4

https://codi-gen.github.io/

https://www.marktechpost.com/2023/06/25/meet-codi-a-novel-cross-modal-diffusion-model-for-any-to-any-synthesis/

https://arxiv.org/abs/2305.11846

ShortGPT

https://github.com/RayVentura/ShortGPT

📝 Introduction to ShortGPT

ShortGPT is a powerful framework for automating content creation. It simplifies video creation, footage sourcing, voiceover synthesis, and editing tasks.

• 🎞️ Automated editing framework: Streamlines the video creation process with an LLM oriented video editing language.

• 📃 Scripts and Prompts: Provides ready-to-use scripts and prompts for various LLM automated editing processes.

• 🗣️ Voiceover / Content Creation: Supports multiple languages including English 🇺🇸, Spanish 🇪🇸, Arabic 🇦🇪, French 🇫🇷, Polish 🇵🇱, German 🇩🇪, Italian 🇮🇹, and Portuguese 🇵🇹.

• 🔗 Caption Generation: Automates the generation of video captions.

• 🌐🎥 Asset Sourcing: Sources images and video footage from the internet, connecting with the web and Pexels API as necessary.

• 🧠 Memory and persistency: Ensures long-term persistency of automated editing variables with TinyDB.

Ship posts and rewrite in style of journalist, author, etc. practice and reading will increase. Grammar and skill will increase, as if mentoring alongside a master. Project: fine tune on editing dataset. Partner with a book or manuscript or other type of business for datasets and editing datasets. Would become a finely tuned small model that performs locally and cheaply. I think there are already sites that do this.

Blog Voice Regenerator

-Toggle button-

Switch to random writer style. Prompt LLM as editor for that writer or style.

• Socrates in the style of modern day editors. Wording and speed and etc changed up?

Something to consider is using custom Markdown syntax to set context values parameters. Initial thought was to use this with Spoke posts, where you could trigger an email from the post body using custom syntax. For instance, emailing all followers of the project to ask a question of them. But, it could also be used as a way for writers to set context inline, but also keep flowing with their main storyline.

Landscape

Since starting this blog, I have not looked at other writing or directing editors to see features. I have before. I will again sure. But I am noting ideas that come that may be useful some day.

https://dreamily.ai is almost what I have been working on. The Wardley map is holding true, which again illustrates the obvious, that a class of tools is emerging in this category. I wonder what can be done to get ahead by predicting what will come next, versus just creating in the space that has been opened. Or, is the best path just to work in the space and get a better sense of how to get ahead from that.

LastMile AI: https://lastmileai.dev/workbooks/clj2y933l000mr0avd2ck42s9

Workbook with some of these features already set up.