Creating with AI 3

Future Telescope 11

Apr 19, 2023

Here we are!

With no further ado, here’s the animated short story video we have been making in our last few editions:

More details in the “What have I learned” section below.

Here’s parts 1 and 2 if you are missing the context:

Part 1: Where I lay out the rules, get Bing AI to help me write the story, and its relevant image prompts. I then generate these images in Midjourney and Bluewillow.

Future Telescope

Creating with AI 1

Today, we’re conducting another AI experiment. We will create an animated short story using just AI tools and nothing else. My hope is that this demonstration will help you find a cool way to tell your stories to the world. Let’s get started. 1. The Rules…

2 years ago · 9 likes · 2 comments · Punit Thakkar

Part 2: Where I use Leiapix to turn the images into animations, Natural Reader to create a voiceover, and YT Audio Library to find the right background music.

Future Telescope

Creating with AI 2

This week, we build on our progress from last edition to make an animated short story. If you missed the last edition, go catch up now, and come back to this one! Done? Alright, let’s begin: 4. Animating the visuals Have you heard of Leiapix? It’s a simple image to animation convertor that takes an image and gives it a basic parallax effect. It’s awesome…

2 years ago · 5 likes · Punit Thakkar

What Have I Learned?

It took me about 45 minutes to make this thing from start to end. That’s not bad considering the high quality of output which would have required you to learn how to make digital illustrations, how to animate them, and how to record a professional sounding voiceover. Not to forget the task of writing a short story in the first place.
We still don’t have a good, low cost, user-friendly text to music AI generator yet. It’s a big opening in the market ready to be captured.
AI voiceovers have gotten scarily good. I saw an ad for “Hostinger” yesterday and it took me about 20 seconds into the ad to realize that it wasn’t a real human talking. Companies are going to go all in on the AI voiceover trend. In my previous life as a marketing guy, it used to cost me about USD 350 to get a professional voiceover done. I would in fact skip voiceovers in many cases because of the costs associated with it. Now that changes. Alas, I am not a marketing guy anymore.
There’s a bunch of text to video AI solutions. Steve.ai is really good at this, but it obviously doesn’t get you all the way for a short story video like this. Works better for corporate explainer type videos. Even Adobe came up with something just yesterday, and I’ll cover it in the catalogue below. For now, I just wanted to share that I edited this video together in Cyberlink Powerdirector on my phone. Video editing is not as tough as it used to be!

I’m curious to know, what have you learned from this 3 part series on creating with AI? Did you follow along and try out your own version of an AI generated short story? Have you been playing around with Midjourney? Did you try out Natural Reader for yourself?

I’m excited to find out what the Future Telescope community has been up to, so please leave a comment sharing your experiences with AI tools so far!

Catalogue

1. AI agents can do their own tasks now: AutoGPT and BabyAGI have been a thing for 2 weeks now, which as you know is an eternity in the AI world. Essentially these are recursive GPT tools. You can ask them to do a task, they will break down the task into multiple steps, and then ask themselves how to do the task. It’s the future.

Check out this thread by Nathan Lands to see examples of AutoGPT and BabyAGI in action.

Read, watch, and learn more about these tools here:

👆This one by Fireship is bonkers. He essentially does the same thing we tried to do with the “Creating with AI” series, but actually deploys code and reasoning to actually make an end to end AI generated video. IT’S NUTS!!

2. This buff Harry Potter characters video shows us the world we are stepping into: A new generation of memes, ladies and gentlemen. Ones which are not limited by templates but by your imagination.

Click the image to view the video. Shame that Substack doesn’t support Twitter embeds anymore. Thanks, Elon!

3. Adobe Firefly comes to Premiere Pro: In my last catalogue about a month ago, I wrote about Adobe bringing its AI tools to the world. Now it seems to be entering Phase 2, integrating it into its Creative Suite of apps, the first one being Premiere.

Read more about it here, and watch the video for more:

4. This Stanford kid connected ChatGPT to his Apple Health account for maxing out his health stats: He suggests one can ask this health assistant questions like:-

"My leg is cramping. What could be causing this?" - "I am training for a half marathon. Can you come up with a training plan given my current activity?" - "What was my last cholesterol level? Is it too high or low?" - etc.

Here’s more details:

I’m happy I tried out a more practical application based approach to AI tools with this “Creating with AI” series. I hope to do more in the future. I encourage you to give it your own spin, find your own way of creating and communicating with these amazing tools. A beautiful world is waiting for us, one where we can share our thoughts and feelings in the most vivid ways imaginable. Are you ready?

A beautiful world, imagined by Midjourney

That’s it for the eleventh draft, see you next week!