Transcripts
1. Introduction: I told you, you could create professional grade
cinematic videos using artificial intelligence. In this current
class, I'm going to teach you how to
harness the power of AI to create videos across any application as part
of cinematography, content creation, posting
it on social media. It's up to you. Such that
by the end of the class, you'll have a clear ability
to create masterpieces, specifically pieces
of video that you could utilize across
various applications. We're talking about creating
videos from scratch. Airing videos, mapping
your own creative vision into an actual video content
that you could produce.
2. Your Project: Your project for
the class revolves around applying the
lesson stop to create your own piece of video
content after which you're going to be sharing
it with the rest of the community for feedback.
3. Exploring ChatGPT's Sora: And welcome back. Now I'm going to show you how
you could utilize HTGPT for video creation.
You heard me right. Typically, you use
HAGPT for images, for prompts, for graphics.
But how about video? In this current class, I'm going to be
showing you how to utilize hATGPT to generate different pieces of
video based content with a short form or long form. That way, you're able to
have an endless pool of inspiration in terms of utilizing AI for
various applications. So how do we navigate this? First of all, AGPT has a
certain integration called SOR, which is the ability to create videos and images through
an extension of HADGPT. Take a look at the
current URL over here. If you have no idea how to
access SRA, here's the link. Simply go ahead and
type in the link in the a web browser of yours, sa.chatgpt.com, and
then you will be prompt to log in using
your HTGPT credentials. Typically, that used to
be present as part of your menu within HAGPT however, now it has been placed as a
separate product by Open AI, and also there's an
application that could be downloaded on IOS systems only. However, through the
web browser access, you have the ability
to utilize SRA across various video and
image based application. The whole point of this lesson is for you to understand,
first of all, how to set up SOR,
how to access it, and how to navigate it before
we dive into the details. So once you put
in one more time, the URL over here
in the browser, you will be able to log into your HATGPT
account and access SRA with the same account
that you have for hATGPT since it's an
extension of HATGPT. Now, you'd realize that you will have a prompt
message that you are using the old SARA model because this is the
web based model. The new model is on mobile
phones, especially IOS. However, for general users, everything which I'm going to be teaching you right now
works perfectly fine, whether for the
mobile application or the web based that being said, once you log in, you will be landing at the following
main homepage. You got the Explore
button over here, which shows you the
different results generated by fellow users. Every single one of those has been generated by
user like yourself, and you have the ability to
either like it, for example, or click on it and take
a look at the details of the prompt that has been used to generate such a
piece of content. For example, this
image over here, if I click Edit prompt, I have the ability to take a look at the prompt
that was used in addition to using it for
my own creation purposes, which we are going
to be diving into. Now, if you take a look at the menu option images,
once you click on it, you are going to filter the explore results
to actually get images which have been created
by fellow users as well. Like I mentioned, these
are the outputs generated by other users who are using the SRA feature
as part of Chat GPT. However, if you don't
want your results to be displayed along the others, you could simply navigate to
your own account details and simply go ahead and toggle off the public sharing
of your results. This is very important for professionals within company
settings in order to make sure that
they are utilizing the features within their workplace policies
and requirements. Then if you navigate
to videos option, you are going to realize that
you have a list of videos, not images filtered from
the entire Explore feature. And over here, you have the
ability to actually just simply view what other
creators have created. If you like something,
for example, you can just simply
click on Edit prompt. You could copy paste
the exact same prompt. You can use it.
You can remix it. We're going to be diving
into this at a later stage. But at this current
stage, we're just simply exploring how SA works. This is the main interface basically for you to
understand what is SOR. Like I've mentioned also, once you are done, you have the ability to access
everything that you have used. For example, you
have the ability to upload your own images or videos and then edit them or create a folder to
place your project. Also, whatever you have liked, you have the ability to
just simply view it. If you have liked the results
from fellow creators, professionals, you
have the ability wants to click on this, you like
it, you're able to see. So you have the ability
to explore and interact with other users
of the AI model. Now, that being said, this
gives you some sort of a comprehensive
overview on SA, right? Now, let's zoom in a
little bit further. If you take a look at the
prompt window over here, you are going to have a lot of important features
that you need to be familiar with before we
dive into the application. And one more time, if you have
no idea how to access SRA, simply go ahead
and take a look at the URL over here, sachagpt.com. It's part of the
Open AI products. Now, that being said, I
have the option to toggle, either shall I create an image or a video
based on my preference? If I switch to image, whatever I input over
here will be utilized to create an image that will be classified within
the images category. If I click on video, that will be classified
within the videos category. Also, I have the
ability to pick up the aspect ratio for
various, for example, we got 16 by nine,
then three by two, one by one, two by three, nine by 16 for vertical format. How about we go for 16 by nine, which is used for many videos such as YouTube, for example. Here you have the ability
to pick the resolution. Now, for the free
hat GPT version and the initial
subscription model, you have the ability to
access the 72480 pixels. If you notice over here,
the higher the pixelation, the slower it gets to compile. However, by 1080 pixels, it's eight times slower. However, you need to upgrade
for the paid version of hat GPT in order to
be able to use that. Yet, from experience,
it doesn't really make a difference because the
720 will get the job done, especially if you're using it to create certain
pieces of content, not the full on
piece of content. If you need some sort of B roll, you need some sort
of a material, short video to inject within your editing process,
this is very helpful. Then you take a look at
the clock over here. This shows you the
duration of the video. We got 5 seconds and 10 seconds which are
more than sufficient for creating content
that could be utilized in addition to
other pieces of content. For longer forms, you need
to go for the paid version. However, for the fee version, you'll able to access these. And finally, you
can take a look at the options, the variations. Either you'd like to simply generate one video
from the prompt, two videos or four videos. Chat GPT will run your
prompt and give you options. And based on that, you can just simply select your best option. Over here, you got
certain presets. These are a set of
prompts that you can simply select, and
based on that, you'll automatically
change the output to fit such a preset. You could tinker with these
for your own experimentation, but they give you
different results. For example, if you
add an image over here by clicking
the Plus button, and then you select a certain
preset every single time, the new preset will
give you a new result. And most importantly, the new addition over
here is the storyboard. The storyboard is a
very powerful feature, and it's one of my favorite
features, by the way, which gives you the
ability to create your own timeline of content, of videos by adding a prompt
within every single frame, for example, let's say,
this is the first frame, you type in something, then
you add in another frame. You type in something else,
a third and a fourth. That way you are compiling your video based on
your own prompts. Now, we're going to be
diving into that for sure for you to be able
to see how it works. But by the end of this
compilation, scene by scene, prompt engineering,
transformation from a prompt to an actual video. You'll have a full on ready
clip that you are able to export and utilize based
on your own preferences. So this gives you a full on
macroview on SA what it is, how it works, what
are the key features. Now we are ready to actually
dive into the application.
4. Getting Started With a Professional Image: Welcome back. Now we're
going to explore how to utilize Sora initially
for generating images. Simply navigate to the
prompt window over here, toggle the option to image. At this stage, you need
to get as creative as possible in terms of
getting the best results. Let your imagination run and use the details to craft a
very powerful prompt. So what I'm going
to do right now, I'm going to recap
some sort of an image, a typical image of someone drinking a cup of coffee within the winter season, for example, and we got the
coffee placed on a table. So what I'm going
to do is create a hyper realistic image of a man drinking a
hot cup of coffee outdoors it is snowy and
it is the winter season. Make sure that the details
are clear and cinematic. This is the area where you add the features that you would like to have within the image, cinematic, artistic,
abstract, it's up to you. And one of my
favorite integrations is the phrase hyperrealistic. This gives Chat GPT, specifically Sra the direction that you would like to
get realistic results, something that looks like
this, for example, right? So let's try this out.
I'm going to change the dimensions to three by
two we'll keep it's fine. We'll keep it as it is,
one by one is square. So let's keep it by three by two, which works perfectly fine. I'm going to go for
two variations, and that's mainly it. Now we're going to have the
e for the image generation. You take a look at the top. Over here, you'll have
some sort of a queue line. Or the actual piece of content that you're
going to be creating. So we have cinematic
winter coffee. It needs about 18% done, and still it needs
a bit more time in order for it to
be done, as well. So and by the way,
it could accelerate, like it jumped 18-30,
which is perfectly fine. But once it's done, the
wheel will be completed. It's very important to
actually experiment with this. By the way, you
have the ability to test this out over
and over again. Now, let's click on this, the actual ready made result, click to view. Here we go. Take a look at both
of these variations. By the way, both of them, they are AI generated. So this is a very
powerful result. It looks quite real. It reflects the prompt
quite accurately. I have the ability as
well to edit the prompt. If I click over here, I can just simply alter
the prompt or I add an image and combine them
together. It's up to me. So this is a basic application in terms of how
you could utilize Sora to help you craft powerful images.
We're not done yet. Let me show you how you could convert such an
image into a video.
5. Creating a Video From an Image: Back. So I have created
the following image. Let me just simply
select one of them. I'm going to go with this image. After doing so, I have the option either to edit the
prompt if I would like to add something or I could simply shift it to become
a video. Here we go. Click. Create a video. Now, I'm going to use the
image as a reference. And by the way, you could use
the image that you create or an image that you upload,
that's perfectly fine. This gives you endless
possibilities. You can use these
features for marketing, for promotion, for content, online, offline, it's up to you. So what would I like
the video to look like? So I'm going to mention, change the image above
to a realistic video of the man drinking coffee and a white wolf runs
in the background. For example, let's get creative. It's up to us to
pick up or pick out the direction of
the actual output. Now, once I change
the dimensions, I'm going to simply pick which area I would like
to convert to a video. For example, if I select this, it will not just it fits
the entire layout properly. If I select this, I have the option to pick which
square area I need to go with. I'm going to keep it with
this, let's say, cinematic. Dimension. Then the actual
pixelation is perfectly fine. Let's go for 5 seconds,
keep it as it is. Then we got two options.
We're good to go. Let's test this out.
Create the video. Now, obviously, we'll
go to the queue again. It might take a bit
longer depending on the complexity of the prompt, but I would like to
take a moment to help you inspired for your
own unique applications. If you are a marketing
professional, you could take, for example, a picture of your product, and you could have
the same approach to have someone
utilize the product as part of your content
creation or if you are using this for Instagram,
Facebook, TikTok, YouTube, whatever
platform that you're at, you're able to take
whatever piece of content, whether an image and
then transform it or animate it turning it into a
professional looking video. So this gives you some
powerful applications, whether you would like to
create a professional image, like what we have
done over here, edit the image through prompt interface,
as I've shown you, or transform the image into an actual professional
looking video. So we have set up
the foundation by learning about how
to create an image. Then mapping this image into an actual video that we
could use in addition to following the same
logic of editing the prompt in order to alter the output that
you are getting. Let's say you create this image, then you create a video. Then for some reason, you would like to conduct some edits. Simply jump back
to the prompt and repeat the process again. That way, you're able to have a list of options,
list of possibilities. Now let's take a look
at the end result. Here we go. See how
powerful this is. So as you can see, we have
the first option over here. Let's take a look at
the output again. Here we go. We got the
first option over here. We have the guy drinking coffee. Perfect, right? However, I mentioned there should be a
white wolf in the background. So as you notice, this
reflects my preferences. So we have the guy drinking
coffee and all of a sudden, a white wolf walks
in the background. This is very impressive. Now, obviously, it might require more fine tuning when
you are generating your own prompt to
get the results that best reflect your own
unique situation. So in this lesson,
we have learned how to transform an image to an actual video utilizing
Sora as part of ChatGPT.
6. Creating a Video From Scratch: Welcome back. Now in this
current application, I'm going to show you
how to create an AI, generate a video from scratch, simply navigate back to the
prompt window, select video. Now I'm going to simply showcase a specific
area of application. Let's say I would like
to create a video about a transformation for a living
room. Let's get creative. I want you to create a video of a dull living
room which is dark. Then the transition takes place where there is a splash of color and rotation
effect which transforms the room into and interior design
masterpiece. Here we go. Let's get creative.
It's up to you to try whatever application
that you have in mind. I'm simply showing you
how you could craft a video utilizing prompt
engineering best practices. I kept the options as is, feel free to tinker with these. You could change it to 16 by
nine or square one by one. It's up to you or simply
change the pixelation. You can go for 720 as well, and then you can run it again. Now, all of these examples are
for demonstration purposes to help you understand how to utilize this for your
own unique application. And the skill set
that I'm passing on to will allow you to utilize artificial intelligence
specifically ChatGPT and SOR in this current case to help you get creative
with your results, whether for creating content,
marketing, product design, graphic design, a
video generation, image generation, a
graphics generation, and the list goes on. Even editing, by the way, we are applying editing tactics through artificial intelligence, best practices, and prompt
engineering, sound knowledge. I've kept two versions for the actual video
output, as well. For comparison, you have the option to select
one if you prefer. However, if you upgrade for the paid subscription
model for ChatGPT, you can get four videos. Now let's take a
look at the output. Here we go. See how powerful this is. This
is very, very impressive. Now, if I take a look at
both of these options, let's start with the
one on the right side. The one on the
right side seems to deliver but not as much. It's quite confusing.
Take a look at the one on the left side over here. This is very powerful. We are starting off with
a basic dull living room. Then we have this turning
transition effect, and here we go. It becomes an interior
design masterpiece. So it took my prompt, as is, to craft such a
great piece of content. Now, I have the
ability once more to actually alter this or
add certain elements. Feel free to experiment
with this as well. Like I've mentioned,
the areas of applications are quite endless.
7. Video Production Through a Storyline: And welcome back. Now we're going to
learn a very powerful, let's call it a leveling
upscale in terms of using AI, specifically SRA as part of ChatGPT for creating
videos as part of a sequence of frames
where you have the ability to get into
the editing of the videos, even the creative
part of the video at a minute micro level. What do I mean by this? If you navigate back to the
actual prompt window, you would see the following prompt window in front of you. Take a look at the final corner or the final right
side of the menu. You would find what we
call as a storyboard. This is a new
addition, by the way. It wasn't the case a
couple months ago. We used to have video, used to have the aspect ratio, the pixelation, the
duration, and the options. So what does the
storyboard feature do? If you click on this, you are going to be prompt back into an actual storyboard
card by card frame. You would still have the
same menu options as before. However, you would
have something as a timeline for a video editor. On the top of the timeline, you'll have the duration of the video which
you have selected. If you change the
duration, for example, you're going to extend it
based on your selection. So how this is quite
powerful or important. You have the ability,
first of all, to either describe a scene
frame by frame or add an image or a video within a specific
scene and edit this. Let's start this with an example to help you
understand how things work out. First of all, I'm going to go through a very basic example. Start the scene
with a man drinking a hot cup of coffee outdoors
in the winter season. That's one. Right? Then I'm going to jump
to scene number two. This is where you
get quite creative, by the way. Use
your imagination. You are selecting scenes as if you are a producer
for a movie, and you are crafting your
own journey, your own movie. You have the ability to delete
a scene or simply expand the actual captions or prompt that you have added
based on this feature. This gives you the
ability to include more details that you haven't thought of
in the first place. For example, I've added a basic prompt for
this specific frame. However, based on this feature, I'm able to expand the actual scene based on
the suggestions from AI. Let's go through it. The scene opens outdoors in the heart of winter with a man bundled up in a thick scarf
and a heavy coat. This is very powerful
and very cinematic. It includes more details that
will help shape the output. And it's up to you, by the way, to add whatever
you'd like to add or remove what would
you like to remove. So it's a very nice combination. Either as a producer
or an editor, someone creating a video, you put your own prompt or
get your own prompt ready and simply click this button over here to help you
enhance it even more. Let's go for the second scene, for example, and
replicate the same thing. And by the way, you
have the ability to add your own image or video as part of the scene to further enhance your unique output. Then I'm going to
mention all of a sudden, it turns dark and there
is a thunderstorm. Now, this is a
very basic prompt. It's up to me to
keep it as is or simply enhance it
with AI as well. So I've given it a
small structure. Then all of a sudden, we're going to transition
from one scene to another with a
bit more details. Then the serene,
snowy scene suddenly transforms as an ominous
darkness overtakes the sky. The previously diffused
sunlight fades entirely. So as you can see the
structure of the prompt, it's adding a lot
more detail to it, which is very
important, by the way. In terms of creating videos, especially using AI models. The more details you add, based on your imagination, the
more unique the output is. So this works perfectly fine. Let's take it one
level further for you to understand
how things unfold. Then I'm going to mention, then in the sky, a great star appears and the
man looks up with surprise. Then, let me expand on this
a bit more. Here we go. Keeping in mind, the
timeline is quite short. We got ourselves 5
seconds over here. So I'm going to just simply
create the video based on these three different frames
for you to see how it works. And this gives you the
ability, by the way, to keep on adding your
own unique footage, your own unique prompt, creating or editing the
videos as you prefer, based on your own
unique circumstances. This could be, for example, an actual product that
you are dealing with, or you are trying
to come up with the idea for a certain
marketing campaign, let's say, as the one we're
doing over here, we are simply
having some sort of a cinematic output for
the video creation part. This will help us understand
how things work out. Then I'm going to shift it to an actual marketing application. So I'm showing you how would
you get creative with this. At the same time, tailor it for a marketing
application for you to understand that there are no certain limits to what
field this works at. Now, let's take a look at
the output over here, right? Take a look at this.
Excellent. Now, both of them, they serve a different purpose. One of them does not
align perfectly with my own view of this initiative. Take a look at this guy
over here, for example. This one works perfectly fine. This guy is sitting
outdoors winter season, all of a sudden, it changes the ambience,
and there's a star. Perfect. So this is great. Now I have the
ability, by the way, if I want to change something, first we I'm going
to select this one as my preferred choice. If you take a look at
the options over here, you have the ability now to edit this frame by frame basis in case you have
changed your mind. As you're going through
it, you might have realized that at a certain
part of the scene, well, I need to edit some. Or you have the
ability to cut it. Remix it if you'd
like AI to create a new video based on certain
changes to see basically, well, what are the options
in terms of the output, and you could select
the best one. You have the ability
to transition between this video
and another video. So this becomes more of some
sort of a transitional piece that you could use or loop the video to have it replay
over and over again. So I've used this to create
some sort of an actual video. Now I'm going to show
you how this maps out to the actual application
that I have in mind, which is for marketing
application. Then I'm going to
mention, here we go. Then the man sees the phrase in the
sky where you are mourning starts as part of a marketing campaign
for my coffee brand. Mugs, for example, hypothetical. Now I'm going to just simply expand this with a more detail. Now I've shown you
how you could use this for your own creative
process, for example, creating content
across platforms, for your social media posts, your YouTube, whatever it is. In addition to more of a focused application
transitioning from the content creation part
to a specific application, which is the marketing
initiative for a certain product or brand in this current
case, for hypothetical cafe. The man on the bench
continues to gaze upwards as the celestial star
shifts and forms a glowing word across the sky
where your morning starts, the next hangs brilliantly the text hangs brilliantly
against the golden, still winter atmosphere, clearly part of a
larger message. Only moments later, it becomes evident as
the phrase fades, leaving the logo for
mugs, the coffee brand. Excellent. See how
powerful this is. Now, let's create the video again and see this transition. Now you're doing
editing using AB. You are actually creating footage using generative models, in addition to editing them using prompt engineering
practices in combination with your
own creative insight as you view the result, interpret the result,
and do or conduct any variations to the result based on your own
unique circumstances. I truly hope at this stage, you understand how
powerful this is where you are able
to edit videos, create videos using solid
prompt engineering practices, compared to the typical approach where you have to record
it with a camera and then take it to an
editing software and try to conduct the edits. Let's take a look at the
options. Here we go. So we have the first option
over here on the left side, and then the one
on the right side. As you can see, on the
left side over here, the guy is holding
the coffee cup, which is basically suggested
as part of the brand, right? You can see, however, that the star doesn't simply reflect the actual theme of the actual context of the
theme that we have in mind. However, the other one
that we have generated, the one over here, for example, it shows that the guy is
enjoying the cup of coffee Where your mornings,
for example, begin, and then you could use this
as part of a messaging. If you notice it doesn't simply
reflect everything as is, which means it's susceptible
to further editing. So I'm going to go back to
the window of the story line. I'm going to mention
the star disappears and the logo or phrase mugs, cafe appears and stays there. I'm going to rewrite this again. I'm conducting some basic
edits as someone using this, for example, for
marketing purposes. I'm going to see how this following prompt is
going to alter the output. And by the way, I have selected two different options
to view this. Let me show you what four
videos will look like. So I'm going to try to change the pixelation. It
might take longer. However, it will help us get better, more
premium results. So I've added this to the queue. I'm showing you how would I go through that process
if I'm trying to do this for a certain
project related to marketing. So you get inspired. You're able to use
this, for example, if you're working
on graphic design, content creation
for social media, digital marketing, online,
offline marketing, whatever it is that
you have in mind. You have the ability to tailor these tools
that I'm sharing with you to help you best reflected on your own
unique application. And the areas where you could use these
are quite endless, whatever industry
that you're at. If you are a marketing
professional, if you are someone working on leadership as a
creative professional, fashion, design, food, whatever
it is, it's up to you. You got the tools to help you level up your skill set,
to help you apply these. Now let's take a
look at the output. As you can see over here, the one on the left side doesn't really reflect what
we're trying to achieve. However, the one on the
right side seems better. Yet the story line, it doesn't seem to reflect
fully what we have in mind. So I'm going to go back
to the storyboard. Then I'm going to remove,
let's say, one of the frames. And by the way, since
it's an AI model, it's very important to
keep in mind that you might have results which are
not exclusively accurate. And then I have the ability to shift the frames as I please. Then I'm going to recreate
this again, two videos. Here we go. Let's take
a look at the output. This might require a
couple iterations for you to get the results that you want in terms of full on a clip, full on a piece of content, or what I would prefer to use it for create a certain piece, standalone piece that you could use as a B roll, for example, or to help you inject it within your own marketing
campaign rather than being used for the full on
content creation process. It's up to you in terms of
how you'd like to use this, but I'm sharing with you
the options that could be good suggestions to help you implement this in
the best way possible. And like I've
mentioned, you have the ability to add
a frame and then drag whatever image or video you have over here
without any dits to it, thinking about this like a transition that
you could implement. Now let's take a look
at the new results. We got the one on the left side, then we got the one
on the right side. Now, both of them
are great options. Take a look at this
guy over here. It seems more of a
realistic representation of what we have in mind. However, it did not reflect the actual logo or it did not reflect the actual phrasing. Why? Because I did
not give it the logo, how it looks like, right? I'm simply creating
a hypothetical case. However, if I go back
and I simply drag a new frame or add the logo to the current
existing frame, then it will have the ability to reflect it on
the finalized out. So this is more of
an advanced tactic. I wanted to share with you
to help you see how you could use prompt engineering
practices in addition to your own creative
insight to help you utilize artificial
intelligence to create video, edit video, and repurpose whatever piece of
content that you have.
8. Wrapping Up: What do you think? Truly hope that you
found the class helpful. If it helped you level up your knowledge in terms of using AI across various
creative applications. It's a job well done
and make sure that you follow my profile for the
latest releases and updates, and I look forward to
receiving your feedback on the current class.
Till the next one.
Engr. Hussein Attié, Entrepreneur I Engineer I Educator
