Transcripts
1. Welcome to the AI Tools Crash Course: Hey friend, are
you interested in what chat GPT is? Mid journey. Google barred these AI tools that you've been hearing
about all over the news, maybe from a friend
or family member. But you're not exactly
sure how to get started, what they are and how
to use them yourself. If that sounds like you, this
is the perfect crash course because I'm going to quickly help you get started
with these tools. These tools actually
aren't that hard to use once you know the basics, and that's what
this course covers. This is not an
in-depth masterclass showing you the most advanced
ways to use these tools. This is a quick crash
course for each tool. In the first five, 10 min, you'll know how to
sign up and start prompting to get great results
from tools like chat GPT, which can write text for you,
meaning writing articles, outlines, social media posts, e-mails for personal
or business use. It's also a great
research tool better than using something
like Google Search or any other search engine. It's much, much more
advanced and it can even edit your text for you, making it more condensed,
changing the style, the format, and so much
more super exciting. And you'll learn how to do all
of that in the first five, 10 min of class. We'll also cover tools like
mid journey that allow you to create AI based
images from scratch. So from a prompt, you just type in some words, spit out an amazing
high-quality image from illustrations to sketches, to graphic designs, vector art style, to
photorealistic images. It's an amazing tool. So click to the next
lesson to start learning GPT in 10 min, or jumped to the tool that
you're most interested in. You there
2. Learn ChatGPT 10 Minutes: Become Super-Productive for Work or Personal Life: First let's get
going with chat GPT. Chat GPT is the spark
of this AI revolution. It's a tool where
you can prompt it. That's a term you're
going to be hearing a lot, prompting,
prompt engineering. This is basically how you request it to do
something for you. It's the text though sentences, the verbiage that you use
to submit to the AI tool, and then it's going to
give you back a result. So head over to Open ai.com. This is the company
behind Chat GPT, or you might hear
the term GP T4, which is the version
that is currently out that we are
going to be using. Click on product
and go-to chat GPT. You'll need to sign up for
an account and then login. Once you log in, you'll see an interface that
looks something like this. Currently there is a
free and a paid version. I'm using the plus paid version which gives me access to GP T4. You can see here when I do this drop-down menu over on the right-hand side of
the page that it gives you the details of GP T4 versus
the previous 3.5 version, which is available for free to you to get started right
now in the future, GPT for might be
the free version, it might be less basic version, and maybe there will be a
4.5 That's for paid members. And you can see that
the GPT for version has more reasoning,
conciseness, et cetera. Whereas 3.5 is faster, the results aren't
going to be as high-quality as a plus
member I can choose GPT for. But if you're not
a paying member, then you'll just the standard
option which is 3.5. On the left-hand side, you won't see anything until
you start doing prompts. But this is our
history of prompts. And you can create
separate chats by clicking the Plus chat button in the top left to organize your chats. So if you're doing
multiple projects, I suggest separating it and
starting a new chat every time you want to organize
it into one thread. It's very simple to use. On the right-hand side, there's a chat send message
box in the bottom right. Here is where we just start asking it to do
pretty much anything, some ideas to get you going. You can ask it
questions for research. You could ask it to
write out any content. And currently chat
GPT is text base. So you're going to
submit text and then you will receive text back. So you can imagine for
businesses this is super powerful for asking it
to write out emails, right out social media posts, right out videos, grips. The process is going to
be very similar for each. It's just going to take
a different prompt. In this video, I'm going to
use travel as our example. And it could do things
for personal travel. We can do research or we can be a travel company and we can
have it do some work for us. So simply we get asked, what are the top
five things to do in Los Angeles for
a family of five? So as of this writes out, you can see some ideas
already populating. This is something we
previously would have done with Google and
then a website with a top five or top ten list
of things to do would pop up and we would have to sift
through that information. But chat GPT has actually put together this list
for us based off of all of the information
that has chat GPT was trained on data on the
internet up until 2021, currently being connected
to the Internet. You'll hear that term, which means and maybe by the time you're
watching this video, it will be connected
to the Internet, which means it's
connected to all of the latest websites,
the latest reviews. And it's probably
going to be taking that information from sites like Yelp or TripAdvisor to spit
out information for us. So know that right
now, that one, not all of the data is
completely accurate. Chat GPT does have a tendency to just make up fake information
from time-to-time. That being said, I think that's sort of an overblown issue
for a lot of things. You can see here that these
ideas are really great for just personal research. But if you're trying
to do research on trending topics and the
latest information, it's not going to be there yet. So this is a great first step. But what's beautiful about chat GPT is we can take
it a step further. We can ask or prompt, create a one-week travel
itinerary with this information. So this is finishing
typing out for us, but you can see that it
breaks it down by day. It gives us a morning, afternoon, and
evening itinerary, filling in the gaps and also including the information
from that previous response What you'll learn is that
chat GPT on these threads, it's using what you've
previously prompted, the information that
it's spit out to you and it's using that for
future prompts. And that's how you can really use it in a more advanced way. You're not just doing one
simple prompt from the scratch, which can start, you can do. You could have asked it. Chat GBD, write out a one-week
itinerary for a family of five with the top five
attractions to visit. And it would have done
something similar. I've found that starting with
a smaller prompt and then expanding on it and re-prompt eating it works a lot better. So this is great if you're
just doing personal research, but this can be used
for a business. Maybe you are a website creating an article or a guide
with this information. Maybe you're a hotel and
you want to have brochures that have information
like this or printouts that you can
give to your guests. You could also take this
a step further and ask, turn this itinerary into a
five-minute video script. And here you can see
that it gives us back a timeline of the video. It has shot ideas for the types of video shots
you're going to get. It's going to have
VoiceOver ideas. And you can imagine
that in the future with tools that are coming
down the pipeline with chat, GPT and other AI
tools that you could plug this script into a
video editing application. It's going to be able to find stock footage that has
what's on these shots, shots of downtown Los Angeles, shots of the Getty Center. It might even have
an AI based voice that will read this for you. Now, you might not want to
use all of these AI tools. You might not want to
do it all with AI, but it could totally
simplify the process of creating content for you
like it's doing right here. Alright, so here it has just
completed that video script. Amazing. What, what do we
need to do with this video? We need to promote it, right? Let's have chat, GPT, a social media post to
promote this video, right? A social media post to
promote this video. And now it's coming back with this social media posts
that includes emojis. It has hashtags, it has a spot for the link for the
video that we'll plug in. All done for us. This is mind-blowing, right? Well, this is just the tip of the iceberg in terms of what
you can do with chat GPT. I want to show you in the
next video how you can engineer the prompts to really fine tune
what you're getting, what you're getting
back from chat GPT. So we're going to continue
with this process, but we're just going to
make it even better. And I'll show you
some more advanced tips coming up in that video. So hopefully this first
video gets the mine rolling. Go ahead and sign up for chat GPT if you haven't done so yet. Again, I'm using the
chat GPT for version. If you're using the
3.5 free version, the results are still
going to be good, but they're not going to be as creative and as unique as
what you're seeing here. So just keep that in mind. Okay, hope you enjoyed this video and we'll see
you in the next lesson. Click over to that
one right now. See you there?
3. Improve Your ChatGPT with Prompt Engineering for Better Images: In this lesson, I
want to show you how to do better
prompt engineering. Basically how we can change what we type in to get
a better response. So far, you've seen
that we've asked chat GPT to create different
types of content, different formats of content. We did a video, we did
a social media tweet. We asked you to just
to put out some text. Depending on whatever
format you want. You can prompt chat
GPT to do that. The next and more
advanced ways to customize our responses
is to adjust who the audiences for
and what the tone is or how we want
chat GPT to respond. What we can do is we can do this from scratch or we can take existing content either from a thread that we've
already created, or you can copy and paste
text from an article or something that
you've written out and ask chat GPT to edit it for you. Let's start from scratch
with a different prompt, but I'm going to add a
couple details for chat GPT. So my initial prompt is URI, high-school teacher taking
a group of students to visit Paris for one
week in the summer, write an email inviting potential students to
join them on this trip. Include the price of the
trip, which is $2,500. Also include a list of five
benefits for participating. Make it exciting and fun. I've included those last bits, which is something you should do when you're prompting if you want anything specifically
back from chat GBT, don't just expect it to come up with ideas or write
things for you. If you want something specific, things like including a list is something you
can ask it to do. Because otherwise it
might just come back with a email with paragraphs. But what it should do
now is come back with a, some sort of list with
those five benefits. Also prompting it as a
high school teacher. And then the last with the tone, make it exciting and fun. That's going to change
how it's written compared to writing for a
group of seniors, adults that are older, taking a tour of Paris. It's going to be a lot different than this email which says, starts out, dear students, we hope this e-mail
finds you well and excited for the
approaching summer break. We are thrilled to announce an amazing opportunity to explore the world
beyond our classrooms. Picture yourself wandering the cobblestone
streets of Paris, basking in the city of
lights and indulging in the rich cultural heritage
and world-renowned cuisine. Now make that dream a reality. Pretty good, right? It has a subject line, something that I'll just
tell you you can do though later on is
you can ask it right, five catchy headlines or
subject lines for this email. And then you can pick and
choose the one that you want. You don't just have to go
with what it writes for you. And of course you can
edit all of this stuff. Here. It's writing
out the five benefits that you get from
taking this trip. Cultural immersion,
language skills, global perspective, educational enrichment,
friendship, and fun. So here it's ending with
to reserve your spot. Please reply to this email
or stop by room to O2. So sometimes chat GBD is just making up what it thinks
you're going to want. And of course we can edit that. Maybe I'm in room
301, not two O2, or maybe we don't need
to have that on there. And it'll just be
an email address. But isn't this incredible? It's written out
this e-mail for us. But the magic of this
is that we can engineer the prompt to get a completely different
type of response. So I'm just going to copy and paste this prompt and
change a couple of things. Alright, so I've written you are an immersive horror
travel company, taking a group of
young adults to visit Paris for one week
in the winter, write an email inviting
potential guests to join the spooky, spooky trip. Include the price,
include a list of five non-traditional things there'll be doing on the trip. Diving into the quirky, unique and scary side of Paris. Make it sound scary
but intriguing. Dear adventurers. What if we told you that beneath the glitz and glamour
of the City of Lights, there exists a spine chilling, darker side waiting
to be discovered. This winter, we invite you to join us on an unforgettable, unforgettable journey, one that will send
shivers down your spine, tingle your sense of adventure. Here's a sneak peek into five of the unconventional spine
tingling activities. We have planned. One, catacombs after dark, descend into the
chilling labyrinth of catacombs where millions of Parisians were laid to
rest centuries ago. Ghost and legends night tour, roam the ancient
cobblestone streets, streets of Paris after sunset, discovering chilling tales
of the city's infamous goes haunted spots and
unsolved mysteries. Pair Lachaise, Cemetery
nocturnal visit. Your mind blown yet,
because this is crazy to me how it can come
up with this e-mail. And that's just based off
of how we prompted it. So you're going to
hear this term prompt engineering and it sounds
confusing, it sounds advanced. But basically all
it is is changing the way you write your prompt,
including more details, including the format you
want it to write back at, including the tone and information that you
want it to include, also giving a role. If you asked it to write a
simple email as a teacher, as a business professional, as a kid, as a
YouTube influencer, the response is going to come
back completely different. And I just encourage you to
play around with it because that's what it's what
you're going to learn best. So this is a pretty
ridiculous, awesome example. Something else to
note is that you can ask chat GPT to help you out. If you're ever stuck, use it as a learning
tool yourself. So I can ask, what are ten roles that
you can play as Chat GPT? In here it's giving us lots of different roles that
chat GPT can play. Educator, advisor,
language tutor, creative assistant,
research assistant. Use it to help you
improve your prompts. Alright, thank you so much
for watching this lesson. I hope by now you
have a good sense of what chat GPT can do for you. If you have questions or you're confused about
anything, let me know. I'm happy to help. But right now, the best thing to do is just play around with it. Think about what
in your own life, personally and
professionally and can you use chat GPT to help with? Because we are
moving into a world where jobs are going to
be taken over by AI. That's a fact, it's
already happening. And so knowing how to use these tools to improve
your own life, to improve the way you do
your job is going to be super important moving forward. And hopefully, this couple
of videos have helped you get a little bit closer
to knowing how to do that. Alright, see you
in another video. Have a great day.
4. Learn Midjourney in 10 Minutes for AI Art Generation: Welcome to this new lesson. In this one, we're going to
be looking at mid journey. So you're going to learn how
to create your own images, photorealistic images, illustrations, and so much
more with this powerful tool. Go to mid journey.com. And if you click on showcase, you can see examples of what you'll be able
to do with this app. From illustrations
and digital art, too realistic images,
you can do it all. In the past mid journey
has had a free trial, but currently, you do have to pay to get started to sign up. You want to click the Sign In button and then join the beta. Mid journey is built on
the Discord application, which is a website and a
downloadable computer app. It's sort of like a community
chat app, similar to Slack, where you can have
different channels or groups are areas
where you can chat and you'll see how it
works in a second. But all you have to know
is that you have to sign up for a Discord account, which you can do by
following the process. When you click on join the beta. If you already have
a Discord account, you'll want to click already
have an account and sign in. If not, you'll have to
sign up for one here. Once you sign in, you can
see your different discord. Groups are sites that you're on. You can see that
I'm in a couple. The mid journey one
is here and here you can see all of the
different channels. There's one for getting started, which is great to check into to see some instructions
for getting started. And then we'll walk you through the process that I'm
going to show you right now to actually create an image at the
little bit confusing. But basically what
you need to do is go to any of these
newbies channels. Don't worry about it. Just go to the newbies channel and
you'll see a stream of comments and images that other people are actually
creating in real time. To create an image, you have to follow
this specific process. Go to the chat box
or the messages and the bottom of any of
these newbies channels. Typing slash,
imagine, and you'll see the different
commands start to prop up here or pop-up up here. And so you can
type in imagine or just click that imagine. And then you simply type in anything that you can imagine, an illustrated cute
cat in a top hat. And similar to what
we saw with chat GPT, you can go super in-depth
into what the prompt is. And that's part of the magic of making the result exactly
what you're imagining. But you can start with
something simple. And you can see
though other chats sort of started to pop in. And what is happening is other people are doing
this at the same time. This is a public feeds so other people can see
what you're creating. But you see what I've done here. It's starting to
populate, render. It's at 31%, 46. And you can start
to see the image. Once it's done, that
chat or that comment will pop down to the bottom as an actual published comment. In here, we can click
on it and we can see the full resolution
of the image. So we've got four
illustrated cats. What do we do with this? It gives us four variations. Underneath this image you see eight buttons are
actually nine buttons. You one through four. And V13 for you means
you're going to be able to upscale or create a
higher resolution image of any one of these. Starting in the top right, It's u1, u2, u3. And then for, if you're happy
with one of these images, you just click u1, e.g. and it will upscale
this first image, V1 through four means
you liked this image, but you want it to create
some variations to see if it can create something
a little bit different. So I say I like this fourth one, but I want to see
what it comes up with it with some variation. So I'm going to click V4 and then I can scroll
down and I can see that it finished
this one job of the first u1 upscaling. And then down below it has my next variation job
that's starting to run. There was one other button. If I scroll up here, it's this rotating
arrows button. And that's just to create
for brand new variation. So completely start
from scratch, do for new ones. So now we have these
four new variations. And from here we
can upscale one, create more variations,
regenerate everything. And you can keep going
and going and going. So I think that first
one is pretty darn cute with its little
pause coming out. So I'm going to
upscale that one. Once it's upscaled, you can either download it by just
simply clicking on it, right-clicking and choosing
to save to your computer. Or you can click Web right here. From here you could
also see variations. But in two, if I click
on the web button, what's going to happen
is it's going to open up the mid journey
site where you can sign in with your account that you've created already,
your Discord account. And here we have our
mid journey library and it's loading
all of the images that I've created in the past
and saved to my account. In here it opens up. And from here we can save it. We can favorite certain
ones that we like. We can share it with people. It gives us the information
and it also gives us some related images that you might be interested
in looking at. And then back on our profile, if you exit out of that image, you can see all of the ones that we have saved to the web. So this is a great, clean way, easy way to save photos
for using later. There's also another
way to create images which I find a little bit more organized and
easier to work with. If you click on the Direct
Messages button up here, and then you click,
and then you click Find or start a conversation. You can type in mid journey
and find the mid journey bot. And here you can see that I have an organized thread with
all of my own creations. So this is not competing
with a bunch of other mid journey users.
It's alright here. So again, we type in slash. Imagine backslash. That is, now let's do a
photo-realistic image of a cat wearing a top hat sitting in the middle of a pile
of fall leaves. The background is blurry. It is a photo shot at sunset with beautiful
golden hour late. This is where you can
get super detailed with exactly what you want
your image to look like. So the main thing is while
this starts to load, to think about when you're
putting in a prompt for mid journey are of
course, the subject. Also what's around your subject, what the setting is, and then the style or format. So is it a photo, is it an illustration? Is it a sketch? Is it digital art is
in a particular style. You can reference other
photographers or periods of Arts. You can be impressionistic
or abstract, modern pop art, Anna
Mae, or manga style. You could reference popular
historical photographers. You can reference camera models, camera lenses to use. And here we have,
this is pretty good. And there's obviously something
about these images that look fake because it's
sort of a magical idea. But in terms of the
quality of the image, the cats themselves look, themselves look pretty
darn realistic. I'm going to do some
variations for this. Variation for light
is beautiful. It captured exactly
what I was going for. If I didn't have the
top hat in there, it would look almost
like a real photo captured at a perfect time
with perfect lighting. Edited really well. You can only imagine what
in the next year to or beyond these tools are
going to be able to do. You'll also notice
that depending on how many people are
using the system, sometimes it takes
longer or shorter for your images to actually
pop up and be created. So these are all
pretty darn good. I actually like this
original one right here. I'm going to upscale
that fourth option. And I'm going to save that
to my mid journey account. Something to note about the direct messaging
option though, is that your images can still appear on the public
mid journey site. They pull images that are great, that are rated well and
they might showcase it. So depending on what
you're creating, it's not a private
private option. You have to pay for the
top tier account on my journey to do what's called stealth mode, which is private. Having all of your
images private. So just be aware of that. Awesome. So that is a quick crash
course into mid journey. In the next video, we're
going to learn how to use our own images
to create cool, fun variations of those images. And also how to use an
image that mid-year and, and created to create a second image or a
third image in a series?
5. Uploading Images to Edit in Midjourney & More Advanced Prompts: In this lesson, I want
to just quickly show you some more advanced things that
you can do in mid journey. Talking a little bit more about prompt engineering
specifically for mid journey, you can see an image
of a little character that I'll be showing you
how to do in just a second. But the first thing is you can actually take a
photo of yourself or really any photo and have mid journey transform
it into something else. This does not always work
when you put your own image. It doesn't look exactly like you yet transform it
into another person. Other AI tools that can do that, that I've seen like Len, I believe it's lenses AI. But with mid journey, if you want to try it out, you have to do this either on the mid journey bot
where you can upload a photo or you need
to find a link to a photo of yourself
that's online. So you can upload a photo to a public facing
website like Imager. Or even if you have like e.g. if I type in my name on Google, it will show up all the
different images that are public and you need a
decently high-quality image. And on YouTube, this profile image is one
that I can right-click. I'm going to copy
the link address. And then I'm going to go
back to my Discord server, either the mid journey bought or to one of the
new comer rooms. We're going to type in. Imagine. Then here's the
specifics you need. The greater than, less
than brackets open, we paste the link to that
public URL, close it out. Chomp, chomp, chomp. Then we say, what we want. This man as a Viking, photo-realistic image with
high-quality quality details. I will just leave
it at that for now. And I got the wrong link. It's not the link
to this page is the link to the image
copy image address. Whoops, Let's try that again. Imagine, while this does not, the other way you can
do that is by uploading to the bot or the messaging bot. So you click the Plus button, we can click to upload a file. Here we can find an image. It has to be a
relatively clean image, and then you send it by pressing Return and that will upload it. And then what you
can do is type in, imagine open chomp, chomp bracket greater
than or less than rather. Then what we do is
simply drag and drop this image into the chat, which brings in the URL
close that bracket. And then we type in our prompt. This man wearing
a pirate costume. Alright, so here we can see
the original null one pop up and you can tell that
there's some similarity. So this image compared
to the other one, it's not too off, but it definitely
looks like more of like maybe a second
cousin of mine, unnecessarily me as a Viking, but still pretty cool. And then of course you
could add all your details. You can have it. You talk about the background. You could say a full body shot wearing a certain
type of costume, fighting off dragons or
whatever you want to say. Then here we're going
to get the next one and you're going to
see the difference in the what it perceives as
the person that it creates. So this one's loading and again, you can see that it's not me. It's not me wearing
a pirate costume, but it's something close. And what I've found is
that you can regenerate it all if you just refresh
everything from scratch. Like some of these
are just off totally, but I could see some
similarities and some but you can just
regenerate it all. And I found that
depending on the prompt, it gets better and better. One of the ones I
think I did cyber punk and it came out pretty close to me versus some of
these other ones. So that is one way that you can just have fun
with mid journey. And I wanted to mention that
mid journey has a lot of documentation for helping
you create better prompts. If you go to your
mid journey account and you go to help an FAQ, it has a lot of
documentation that will help you click on
the user guide. And here there's a ton
of things that you can look into, like parameters. So these are things that
you can add to the end of your prompt to change different
things like aspect ratio, and it's highlighted
in what you would add. So dash, dash AR, 16 by nine is aspect changing the aspect
ratio to 16 by nine. And you can see some examples of different aspect ratios here. Just go through these
on your own time. I don't want to waste your time talking through
each of these quality. Quality is going to increase the detail and
quality of the image. It takes longer, but you
can increase the quality. Stylize is another one
that's kinda cool. Dash, dash stylize. And then you put
in a value 120500, different ones up to 1,000. And you can increase the style and style low, medium, high. And you can see opening
up these images, the difference from
a lower quality to a higher style or lower
style to hire style. See how the
hairstylists like much more deeper, much more detailed. And then beneath that, we have some advanced prompts as well, where you can actually
combine images. So this talks about how
you can upload an image, which I just showed
you some examples, but you can actually
upload multiple images and combine them. One last thing up under
the Getting Started, under next steps under prompts. This is a great description of the different things that
you can add to a prompt. So down below there's
this list right here, talking about subject, medium, environment, lighting,
color, mood, composition. And it gives some examples. These are all things you can
play around with to come up with great unique
images with my journey. Alright, I hope you
enjoyed this lesson. In the next one, I'm going
to show you how to create a consistent character design
from one image to the next
6. Creating a Series of Images in Midjourney with the Seed Prompt: One of the current
issues with mid journey, if you're trying to
design graphics or images that multiple
graphics and images that you'll be using
in a series for a website, for an illustrated book
or something like that is that if you repeat
the same prompt, it will come up as
a different image. Or if you just slightly
change the prompt, it will bring up something
completely different. So how can we make sure that the images that we are
creating can be consistent? Here's a couple of
ways you can do that. First, let's start by
creating a character. So what I'm going to do is
imagine an illustration of a young boy about to
go on an adventure. He is wearing a backpack. Show the full body
character design. And we'll just leave it at that. Cool. So now we have these
illustrations. These look pretty good. I'm just going to go
for one right now. You could adjust it, you can create variations, but what I'm going to do is upscale this one
image right here. So this one looks pretty good. So our next step would
be to type in imagine. And then I'm going to
put the open bracket, close bracket with this
image inside so you could just simply drag and drop
it there with the brackets. And then we're going to say
this boy walking away down the sidewalk about to
leave on an adventure. He looks excited. Now while that one loads, that was the basic way to do it and it's not
going to be perfect. It's not going to be that great. Sometimes the images, they
look a little bit different. They don't look exactly like this drawing might be
in a different style. So how can we prompt my journey
to maintain that style? There's something
called the seed. It's sort of like
the reference code to a specific image and style that each image created has to get it right now it's
kind of a confusing process, but what you have to
do is you have to react to this image. And the way you do that is by right-clicking add reaction. You're not going to see
envelope right here yet. You have to click
View More and then do an envelope and do the
single envelope reaction. What happens when you do
that is that you get a, you actually get a message
that shows and you can see here that shows
the seed number. Okay? So keep that in mind because now we see the results
of that original image. And you can tell
that this character here doesn't look like the one. They're close, but
they're not perfect. What I'm going to
do is I'm going to repeat that process
with the seeds. So imagine we're going to
still take our original image, drop it in here, and we're gonna do the
exact same prompt. But we're also
going to add dash, dash seed, and then the number, the seed number right here. Simply put that in. And it should do a better job at matching the style and
the character and the person because it knows
we're referencing that exact one rather
than what it's doing before is it's
imagining a brand new image. It's trying to analyze
this image that it created and then put it into the new scene
that we prompted it. This one, it's
actually referencing the back-end data used to create that original
image that we liked. And while there's loads,
I will say there is one that doesn't look
exactly like it. And that is partly my mistake, but I've found that
it will do this. Sometimes there's
one or two that doesn't look exactly like it. But however, the other
three look much, much, much more similar
to this one right here. So this character right here at the top-left,
bottom-right, right. So 12.4 look much more similar. What I should have added was an illustration
because that would have been a little bit
more of a specific prompt. So including the style that
you want it to return. Because if I would have put photo-realistic image or
any other style of artwork, it would have come back
as something different. But maintaining the style that you prompted originally for your original character
is super important to include in your
follow-up prompts as well. And now we have a much better representation of this by adding
that illustration. The style is very similar. A couple of these didn't
follow the prompt of walking away down the sidewalk. You could add that. Walking away from camera. You could add all your details, change the things
like the aspect ratio or whatever else that you want. Here we have a much more
consistent character than before. Play around with it. It's not perfect yet. I'm sure that as mud
journey and improves, there's going to
be techniques to improve the
consistency of images. Have fun with this. I hope
you enjoyed this lesson. Have fun with mid journey. If you have questions,
let me know. Otherwise, we'll see
you in another lesson. Bye
7. Learn Google Bard in 10 Minutes: In this crash course, you're
going to learn how to use Google bar and Google Bard
is very similar to chat GPT. So if you have
watched that section, this is going to be
somewhat repetitive. It's just the format of the
tools a little bit different, and also a couple of the
functions that I'll mention in the following
minutes basically. But you can just go to bard.google.com and sign in
with your Google account. Up until now, it has been sort
of a Beta waitlist thing, but I believe now
anyone can access it. They have a Google account. But that might not be true
for every single country. Once you login, you will see something
that looks like this. On the right-hand side,
you have your prompt box. You also have a microphone
if you want to use that, to just do it via verse voice, you have a dark theme which
is pretty cool to that, a little bit easier on the eyes. So I'm going to switch
to that for now. You have your past
barred activity over on the left-hand
side, support, etc. It's still note
that there might be some mistakes or things
on factual results, so you have to be aware of that. Currently, it's just text
responses similar to chat GPT, but they are adding an
option for an image creator and also a tool where you can submit images as part
of your prompts. So let's just do a basic one. Make a list of ten
toddler craft ideas. This is great for me personally to help me take care of my kids. Or if I was blogging, writing an article, creating
a video about craft ideas. And it pops up here
with our option. A couple of differences
that you will notice with Google Bard is that it has a draft
option or version options. So here at the top,
there's multiple drafts, so there's actually three
that pop up for every prompt. You can also regenerate
right from here. So it keeps it a
little bit cleaner. If you're using the 3.5
version of chat GPT, you won't have all
the formatting that Google bard results. And P24 is pretty good with bullet points and lists and numbering and
things like that. But Google barred already
for the free version has it comes back
with nice formatting, bullet points, outlines
and things like that. So this is pretty cool. It comes out with great ideas. Down below two, you can either say it's a good
response, bad response. You can also export it, which is very cool. In chat GPT, you just have to currently just copy
and paste the text. But here we can export to an
e-mail or to Google Docs, which is super cool. The other thing you
can do is click Google it and it will open up a search for this idea. And here you can
just find directly from Google barred their
search results and go kinda compare notes if you want or if you
want to just kinda use it in conjunction
with your regular search, similar to chat GPT. We can then prompt it for
our next texts that we want, right, a five-minute
video on the first craft. And so because it's AI, it's smart enough to
know like we don't have to specify the
paper plate donuts. It will know what we mean
by that first craft. So here it comes back with more of an outline
version of the video. So as the introduction,
the materials needed, the instructions, some ideas for variations
and a conclusion. It also gives some tips for what you might want
to include in this video, which is kinda cool too. We could also prompt it from the beginning with this idea. So we could say, write a five-minute
script for a video, teaching kids how to
make paper plate donuts. Make it fun and exciting. The audience is
parents of young kids, as well as kids ages
three to seven. So when you're writing your
prompts, you remember it. You're not just asking
for the content you want, but you are asking for
the style that you want it to come back
and you're telling it the audience that
it's writing for. And this is what we call
prompt engineering. Changing the prompt to get
exactly what you want from it. Google barred doesn't
do as good as what chat GP T4 comes back as. It's very similar to
that earlier prompt. But what it does, it gives us some additional ideas for lines that you might want to
save throughout your video. Can you make a doughnut
with sprinkles? Can you make it done
it that looks like a cat, so that's kinda fun. It lets us do a little bit more of the work than just
doing it all for us. But it's very, very
thorough as well. So this is a great tool for
making your job easier, but not doing it all for you. We could take this
paper plate donut and change up the prompt, write a 500 word article
about this craft. And then we can
paste that in here. It has written that
out in more of an article format
with instructions, tips, ideas for making the paper plate
doughnut even more fun. And it kinda concludes it. I always make sure to
check out the variations. You can kinda combine the
different variations too. Get exactly what you
were looking for. Or there might be one that's in a different format that is more of what you
were imagining. So that's how you
use Google barred. Currently, I find chat GPT
to be a bit more advanced, but we know that Google bar, it has the Google name
and company behind it. And it's only going to get
better and better and better. So I wouldn't write it off yet. But at the time of recording, I would do most of my work
with chat GBT 3.5 or four. If you are a paying member, I just find it to
be a bit better. But hopefully in
this crash course, it helps you understand some of the capabilities of Google Bard. The other thing
that Google barred has that chat GPT
doesn't have yet, is it's connected
to the Internet. So what I could do, sorry for keeping you
for just a second, but I wanted to mention, you can type in things like, what are the top five
restaurants in Claremont, California where I live? And it will give it real-time results based
off of current data. So it knows it looks at different review sites and probably Google reviews as well. And it will come back
with these results. And chat GPT, since it's not currently connected to the
Internet, does not do that. However, as I've mentioned, they are doing that
currently they're planning to connect chat
GPD to the Internet. So current reviews and
Internet information will be used in its
responses very, very soon. But that is one When, I guess for Google barred
at this current moment. Alright, I hope you
enjoyed this crash course and we will see you
in another lesson. Bye
8. Learn Dall-E in 10 Minutes to Create AI Art: Welcome to this crash
course on dolly. Dolly is an image
generation tool similar to mid journey and it's put out by OpenAI and it's
a very simple tool to use. So the pros of dolly or that
it's very simple to use, you actually get free
credits every month as of now to test it
out and play around. The cons. Are that the
quality of what you get back in terms
of your images, the details, just the
professional quality is not as good as what you can get in a tool like mid journey. But let's try it out
and I'll walk through. It's very simple. Go over to open a i's website, go up to product and then dolly tool and
then choose, try Dalai. If you already have an account, then all you need to do is sign in in the top-right corner. You will see that you have a
certain number of credits. Right now I have 79 left and that's because
I've purchased them. But you do have some
credits to start out. At the top of the page, you have a search bar where
you put it in a prompt. You have an option to
upload an image to edit. Then you also have some samples which are
great to look out, look at to see the examples of the prompts that were
used, these images. They pick some of
the top ones that have come back from the tool. And that's makes sense. But I've found
that it's just not as easy to get a
great quality image, especially more of the
photo-realistic images with Dali. However, let's go ahead and try. I'm going to show
you a basic prompt, then we'll talk through
some more advanced props. So a basic prompt would
be a fox in a forest. It's going to load through. On the right-hand side,
you can see the history of images that you've generated. And you can see that I've tested out a few different
ones over there. And so here we have some basic
Fox images that look okay, especially when they're smaller, when you open them up like this, and you can open them up
and then click through. You start to see
that they don't look as high-quality as what we saw on mid journey
in that section of class if you've watched that. So what we need to do is add some more detail to the prompt. And again, watch the
mature any lessons because you can use the
same prompt ideas here. So what I'll add now is a
photo of a fox and a forest. Professional wildlife
photography. I'm going to even
put the focal length of the lens being used. So I'm going to put in
300 millimeter lens, which is like a big
telephoto lens where you can zoom in far into wildlife. I'll put beautiful lighting, sunset back lighting,
and photo-realistic. Now let's generate it and we'll
see what the results are. Once you get your four results, you can make variations
of those images. So here we have photos
that look a little bit more artistic right
here from this view. Look better. Beautiful sunset
that backlighting, hitting the for in
this one but the eyes, the face of the fox. It just doesn't look great. And that's where Dolly
fails as of now, I'm sure in the future the
tool is going to get better. But here what we can do
is create variations. So what we can do is just
click on an image that we like and then choose
to create variations. And then it will create
for images based off of that photo. One cool thing about Dolly, different than mid
journey is it has a bit of a photo editing aspect or image
editing aspect to it. So here we see just doesn't
look amazing, right? So let's try a
different prompt to see what we can come up with
that might look better. And I think some of
the more illustrations the 3D renders, that kind of stuff
looks better with Dali. So what I will do is a cute 3D render of a
penguin in the South Pole. We'll just do that. Okay, so we've got a few
options that might work. Let's go ahead and
click on this one. The eyes don't look good. Let's click on variations
to see if we can fix that. Get a new variation. You can see tips down
below that pop up. Alright, so maybe our
penguins just going to be closing their eyes
and that's fine. We can start from here and I'll show you how
to edit a photo. So of course you can
download by clicking the Download button
or we can click edit. One of the things that
we have is called out painting or adding a
generation frame down here. So we can click this Plus frame and now we can add
to this photo. So say we want to add more context to the
left of this image We can click there and
then choose Generate. And we can even type
in what we want. A polar bear on and
on the South Pole, swimming near an iceberg. 3d rendering of generate. You also have tools
here to add an image. So if you have graphics or
images you want to add to your generation
or to your image. And then also an Erase tool where you can erase something. So here we have now
a polar bear and it has variation so
we can click through. So this one followed
it with the swimming. This one's kind of cool though. I kinda like that one, so I'm
going to accept that one. We can erase things. So say we don't like
this cloud or something. We can click the eraser and
we can erase that cloud. And now we would have to
add a generation frame, put it behind, or we
can extend it here. And we'll say of three baby
penguins playing with a fish. And it should fill
in that Erase spot and then continue
the generation. You don't have to generate more. You could just
generate what's behind that Erase spot and
it should do it. Alright, so here's
our variations. So we're not going
to be competing with DreamWorks or Pixar yet. But okay, not, not
terrible, kinda funny. Okay, So let's just
accept that. From here. We can download the image
if we want to our desktop. Cool. Alright, so let's go back. So here we have all
of our variations. You can see this was a different prompt 3D
renderer of acute penguins, solid light blue background. Then I added the details of
different applications that create 3D renders to add
some of that detail. A couple of others
that I actually liked. So here is Yosemite Valley and the style of vector artwork. So this one came
out pretty cool. I thought here was a
high-quality photo of a camera. Again, it's not what you
would get back with the same exact prompt
with mid journey. Still, it's a good tool
to know how to use. You can see that some of these other images
look pretty good. So let's go ahead and
do, try to copy this. A van Gogh's style painting of an American football player. A Van Gogh style painting, a woman playing guitar. So here we have these more abstract images
actually look a lot better. I'm going to create a
variation for that. So right now, here's an
example that pops up with the tip stained glass window depicting hamburger
and french fries. That's great. So always think about not
only the subject matter, but also the style. And if there's any reference to other people's style of
painting photography, the specific format,
3D rendering, a painting, a photo, a sketch, pencil, sketch, watercolor,
the medium of the artwork. Always think about all
those things when you're creating a prompt and definitely check out the mid
Journey section if you haven't
watched that already. So this is Dolly, I hope you enjoyed
this lesson and one-star li starts to get a
little bit more advanced. All create some
more lessons on it in terms of how to properly improve your prompts and
things like that to really get high-quality
images from Dolly. Alright, Have a beautiful day and see you in another video.
9. How to Add These AI Tools to Your Workflow: Now that you understand how these tools work,
how to get started, let's talk a little bit
more about how you can implement them into
your existing workflow. This is gonna be
different if you are just using AI for personal use. If you work for a company
and you're trying to use AI tools yourself to
assist you with your job. Or if you run your own business. If you're a creative, if you're a YouTuber, content creator, artists,
graphic designer, you're going to see a lot
of ways you can implement AI into your workflow
in this lesson, There's three basic
options for using AI. The first is during
the research phase. So this could be everything
from idea generation, coming up with ideas for your
content using chat GPT just to brainstorm topics for videos, brainstorm
article ideas. You can actually do research
on the subject matter. So not having the tool do the work for you in terms
of writing the content. But you could research a subject
and then do it yourself, similar to what you would
probably be doing before. Now, by doing a
quick Google search for a subject that
you're not an expert in. You can also use it
to generate keywords, keywords that you can implement throughout your
content publications. So this is more geared towards, towards people who are
putting out content, written, current
content, audio, visual, whatever type of content
you might be using keywords throughout that
written content in your titles, in your tags and keywords
for your content. So you can actually use chat
GPT to basically ask it to give you keywords related to your topic, consumer analysis. What does your target
audience like? What are they interested in? What is currently trending? And as Chat GPT connects
to the Internet, you'll get even
better data on this. Google barred currently is. So you can do that
easily with Google barred and then
competitor analysis, see what other people are
doing easily with chat GPT. So something that I've been
doing as a YouTuber is checking out using Google barred coming up with
list of competitors, coming up with ideas for videos that they've
done that are trending, that do really well. And then trying to
figure out where, how I find my place as a
YouTuber using that data. The next phase is
content creation, and this is what a
lot of you might be interested in doing. Things from writing
articles or documents, writing emails, scripts,
social media posts, books. A lot of that boring stuff. If you run a business
documentation, contract drafts,
all of that can be done initially with the
tool-like chat GBD. And of course you're going
to have to edit them. You know, your audience, your brand, your
business, yourself. And most of us aren't
going to want to just use exactly what
spit out from chat, GPT, or any of the other tools. But it can do a lot of
the busy work for us. And then there's the tools
like mid journey or dolly or any image generation tool that can create
social media images. Images for YouTube thumbnails
creating graphics for ads, any sort of design or
images for print cell. I've been seeing a
lot of people using these tools to create patterns for that people sell as
backdrops for graphic design, for wallpaper, prints,
for prints on t-shirts, mugs, you name it. You can create those
graphics within these tools, creating website content
images for your website. And then of course, using
the images that you create for videos
that you make too. Then lastly, you have editing. So all three of these
phases can be done with AI. Or if you have existing content, you can edit it with
a tool like chat GPT, condensing long-form content
to a short, shorter format, checking for grammar,
spelling and punctuation, fixing closed captions is an example that I've been doing. Youtube automatically
transcribed our videos, but the capitalisation, the
punctuation, is all wrong. Oftentimes, even
words are incorrect. And chat GPT because
it's intelligent, it can tell what the correct
word should have been. And it automatically, It's incredible how
quickly you can edit. And I'm going to show
you in the case study, edit captions to make
them look proper. It's insane. You can quickly change the tone of a anything that's written. So if you have something
that you read them, but you wanna make it funnier if you want to make
it more serious, more professional,
apologetic, more respectful, whatever it is, you can quickly do that and then translations. And we've been seeing
a lot of proof that the translations from
chat GPT are even better than what Google
Translate has been doing because it does it in a more
humanistic natural way. That's not going to be
true for all languages, but it's something to test out. And obviously any kind of
content that you have ready, you can just plug it in, throw
it in, have it translated. There are limitations
on the length of texts, but it's great for
doing it either one at a time or if it's
short form all at once. One idea going back to the keyword generation
is you can actually go back to your old articles on a website or any sort
of landing page, web content, your YouTube video description,
your YouTube titles. You can edit them to make
them more keyword friendly. And you would just
write a prompt that asks to either if you have a list of keywords to use these keywords
throughout the content, or simply to make this
title more SEO friendly, to make this article more
keyword SEO friendly. And that's one idea you
can do with editing. Something that you would
have to previously pay someone lots of money to go through all of your
website content. You can do yourself within
a matter of hours or days, or how outsource it to someone
who can do it for you. So these are just several
ideas for how you can use these AI tools in
your existing workflow. If you have other ideas, let us know in the comments
of the class, we'd love to see what you
are using these tools for to have a great day and we'll see you
in another video.
10. Ethical Considerations and Limitations of AI Tools: As you start using these tools, It's very important to have a conversation about the ethics of AI and how you can
use them properly. First, let's talk about
the limitations of AI. All of these tools are
getting better and better. And who knows what the world
will look like in a year, two years, let alone 510 years. And I think it's great
that you're in this class. We're learning these tools because It's going to
help us stay ahead of the curve and know how to
use these tools properly. Right now, these tools, especially like chat, GPT, or Google Bard, They have some inaccuracies
in the responses. They sometimes present false
data, false information. It can appear like it's coming back as a fact when
in fact it's not. And these are things that
you have to be aware of. Now, I do think that this is
an overstated fact about it. This is obviously a true thing that not all of the data
is coming back, correct. But in fact, I've found that a lot of it is very,
very, very accurate. That doesn't mean you
don't need to edit your work or the
work from chat GPD. It doesn't mean you should
just copy and paste without looking at it
and doing fact-checking. But it just one of those
things that you're going to hear continually from people because it catches a good headline that
it's making mistakes. Of course it will, but
it's just something to be aware of when you're
creating content. And then there's other
limitations as well. The uniqueness of the responses. Yeah, it's gonna get
better and better the more creative
your prompt is, the more input data
you provide for it. It can change the way
that it responds. But imagine a world
where five years ago everyone was writing those
top ten listicle articles, Buzzfeed style articles. And still people
write those today. And they're not great content, partially because they all
end up sounding the same. And that's where
AI is right now. If 100 pupil prompt it with
the same idea to write an article on the top ten
places to visit in Paris. Those hundred articles are going to be very, very similar. And so as a creator, especially a creative
person, business, you need to find a way
to use it as a tool that assists but doesn't
do all the work for you so that you still
have a personality. Because I find that
a lot of times, even if you put in the style prompts and the tone prompts, it lacks somewhat
of a personality. You ultimately have
that connection to your audience and
AI is not going to completely replace
you right now anyways, where these tools stand. But I definitely
think as a creative, as a content creator,
if that's what you do. Because there's so much content
being put out there and every topics being covered
1 million different ways. And there's really
no lack of content. It's more important than ever to have a direct connection
with your audience, which you uniquely can provide. The size of the limitations, the ethics are a more
important conversation. So there's four points that I wanted to just briefly cover and you should talk about
them with your team. If you're using these tools as a team or if you're
using them yourself, come up with a way to be ethical in your approach
of using these tools. So there's transparency,
data privacy, copyright, and then bias
and discrimination. If you're using these tools as part of a business service, where you are writing content, where you are editing
content or anything else where you're using
these tools to assist you. And the end result content is something that you yourself
are not creating yourself. It's super important
to be transparent about that when you are
pitching your services. And maybe this is
just a moral compass, a line in the sand
that I believe in. But I don't think it's ethical for people to be
pitching themselves as copywriters or
editors or book writers, and then just using these
tools to do it for you. Second, we have data privacy. This is an issue
that is a big one. And basically the issue is
that tools like Google, Chat, GPT, all of these, they're using information that has been scraped
from the Internet. And because it's AI, it's not like someone's choosing which article to
use, which data, which book, which
content to consume, which then makes up
the AI mind that it can spit out in a response. And it might in fact receive some private data
that should not be there. So that's why it's
important to read whatever your responses are and you're
putting out into the world. Because if it is actually
violating someone's privacy, that could be a huge, huge deal. Similarly, you have copyright, it scraping all this
information from the Internet. And there's going to be
copyrighted information that it's using to formulate
its responses. So you have to be careful about. So again, you just
have to be careful about what you put out and try to have a step in your process to verify if what you are using has any
sort of copyright issues. And lastly, bias
and discrimination. Similarly, because it's using information that humans wrote, there might be bias and
discrimination in that. E.g. there might be articles or books that have inherent biases. And of course that's sort
of a blurred line as well. But for some people
there might be biases in some information
on the internet. I think we can all
accept that these tools, if they're using that
as part of a response, can be a problem too. I think most of you watching
are likely good people who have good intentions of using these tools
to assist you. You're not monotonically trying
to figure out how you can take advantage of the system using these tools
no matter what, even if it's hurtful
to somebody. So that's why I trust that you're going to be
doing in the right way. But I do think it's important to think about all of
these ethics and limitations as you start using these tools in your
job or business
11. Real World Case Study: How to Save Hours as a Content Creator with AI: Now that we know how
to use the tools, I want to show you a
real-world example of how I am using them to help me
with my content production. So I'm going to
show you how I use this to create a YouTube video. Everything from
researching topics, outlining the video,
scripting the video, and not just using
what spit out, but being able to
properly prompt, edit those responses to get
exactly what I'm looking for. Create the graphics for
the YouTube thumbnail, writing out the information
for the titles, description, tags, and fixing
captions at the end of it. So let's jump in. This is going to be a little
bit of a longer video. But hopefully it'll give
you some great ideas of how you can implement these
tools in your workflow. So I've been putting
out YouTube videos for quite a long time on
my personal channel, and recently a couple
of buddies of mine, we started a new channel, photography and
friends, which is the brand we have for our
photography community. We run a lot of live
streams on here, but we don't put out a lot of content outside of
the live streams. And I think it would be
great to be able to put out some more great trendy
types of videos, specifically shorts,
which is the format that a lot of people are using now to grow their channels. So I'm going to walk you through the entire process
of what I'm doing. I'm doing this from
the beginning. I haven't prepared this
because I didn't want it to sound like it
was too over-prepared. I want you to actually see
the exact process that I go through as a creative
to do this kind of work. So the first thing I would
do if I was starting from scratch or even where
I'm at is do research. I already know some ideas for what I could create
a short video on. But I think it's worthwhile
to do a little research too. So for this, I'm going to use both chat GPT
and Google Bard. And you'll see the difference. The first thing I might ask is, who are the top ten YouTubers
in the photography space? Because chat GBT at this moment is not
connected to the internet. It's going to choose
photographers up until September 2021
that were popular, which generally will
still be popular. But it's getting connected to the Internet right
now as I speak, they're doing beta
trials of that. That's why I'm going to
use Google Bart as well. I'm going to do the same exact
prompt and we're going to see what the YouTubers are. So here we have some though. Peter McKinnon, Jared Poland, Tony Northrop,
Thomas, he didn't. Mango Street. This is all great. And as a creator, YouTuber in this space, I know that this is true, although I also know that some of these YouTubers have kinda like dropped down in
terms of popularity. Others have sort of popped up. So here let's go to Google Bard, we have very similar F stoppers, the art of photography, Thomas eaten Mango Street. So very similar, but
some different ones. What I can do to take this research a
little bit deeper is how many videos have each of these YouTubers
published in 2023? Let's see if it knows
how to respond to that, because I want to
know which ones are more popular right now. And here we have
a list of videos. So f stoppers has a
lot of photography. They all have generally a
lot about the same amount. It's kinda funny that
it's on a round number. I'm not completely sure if
that's going to be accurate, so I would have to verify it. But it's a good starting point. And Google, google part chat
GPT wouldn't be able to do this with the numbers as of now until it's
connected to the Internet. What are the most
popular YouTube videos for these YouTubers? We're gonna do the same
thing with chat GPT. Stopper is how to take
a stunning landscape, landscape photo. The Ultimate Guide to
composition, a year in the life. How to edit your
photos Like a Pro, how to shoot stunning street
photography. Okay, cool. We could even do this. We could pick one of these, say, let's pick Let's
pick Sean Tucker. Was he on this list as well? He's not on this one. Oh, yeah, there he is. I think he was on
this one as well. And here we have. So this is cool. They came up with a
few different options. Three top videos, we could ask for the top ten
videos from each YouTuber That's cool. Okay, so here we have like
five photography tips. You need to know cinematic
B-roll tutorial. I'm looking for something that's more related to photography, photographing the
Northern Lights, I'm just looking for
something that pops out five portrait
photography mistakes every beginner makes
this is awesome. That's a really good topic. So what I could do is use this
topic and ask the tool to come up with something similar that's kind of on
the edge of just copying. So I don't want to
necessarily do that. Let me look at the Bard. I'm going to ask, what
are Sean Tucker is most popular YouTube
videos in 2023? I should have asked for a
list of like five or ten, but how does shoots doing
street photography, the best tree for
diary camera as the best street
photography lenses. Street photography does a
lot of street photography. Best street photography cameras through photography lenses. This is a topic that my
students always ask us. And so using my
personal experience, this is a topic that
stands out to me. So I'm just going
to go ahead for the sake of time,
use this topic. But this is kinda gives
you some ideas for how I use this tool is
to do research. So what I'm gonna do is
write a list prompt, write a list of the
top five lenses for street photography and see what it comes out with
with just that prompt. I don't think it's going to
come up with the best list. And because again, chat GBD is not connected to
the current internet, we're gonna get some
potentially older lenses, although I like what it's
putting out already. It is putting out just the lens itself, not specific brands. So that's good because
it might be outdated. But if I want specific brands, I could ask for it. Okay, So this is cool. This gives us some brands,
some specific lenses. I could ask. What are the top three
street photography lenses for each brand? Nikon, Canon, Sony, Fujifilm, my missing any of
my favorite brands. These seem to be the
most popular ones that my students have. And this is good
because we're getting some initial information
which makes me think, well, this is a random list. I think there'll be
more practical to have for each brand. So this is cool, this is cool. I like this list,
I liked this list. We got this good list of general lenses here,
which is nice. I'm gonna do the same
prompt for chat GPT. See what they come up with. What are the top three street photography lenses
for each brand? And this is on the free
version of chat GPT 3.5. I just want to show
you what it can do. So it has the Nikon 3,550.28. Let's see what bard
was 355-020-8208, and then Cannon was 3055, 0204. 3055, 0204. Interesting. So same, different
order but the same. So this is good. If I'm not a total expert, I could probably rely on this. I would probably go
do a little bit of research myself just to confirm, but for the sake of time, I'm just going to
take this list here. Looks like it's doing well. It's pretty good on both. I kinda wanna just use
the list right here. So I'm going to
ask Google barred, write a video script, 62nd video script
about this topic. I'm just going to
include what are the best street
photography lenses? Actually, that's probably
going to screw it up because I want it to use this list. So I'm going to include
that in the prompt. Use these lenses. What I should've done was
just simply put, right, a 62nd video script about this. I've found that
the video scripts on Google barred aren't as good. So what I'm going to do
is go over here and say I'm just going to ask it right. A 62nd video script
for a short form. Social media. Real about this topic. Using this information. You provide it. Because it has that information, it should pop it up over here. Discovered the perfect
lenses for capturing the essence of
street photography. Nikon, Canon, Sony, Fujifilm, the top brands with
lenses that brings street photography to life. For Nikon and the easiest, the Nikon AFS night core
35 blah, blah, blah. Okay. So only is picking one lens. So I'm going to rewrite
this prompt. Let's see. So we're not getting
exactly what we want. That's okay. We're going to re-prompt, write a 62nd video
script listing the top lenses for
street photography. Use this list. Brands. Nikon, Canon, Sony,
and Fujifilm. Use this list and include all three
lenses for each brand. Alright, let's see
what that provides. What's will you provide? Chat, GPD, God's. Let's explore the top lenses for street photography across
popular camera brands. For Nikon enthusiasts, the Nikon 3,051.8 is a stellar choice. Cannon user swear
by the canon 50. Okay, so that's good. It might be limited
on the timeline. So let's go ahead
and rewrite this, but include the other
two alternate top lenses for each brand has a remainder. Here is the list I
want you to use. Similar. See if it does. Great, perfect. So it's including the Nikon. All three Nikon's. Great. Awesome. Okay. So it's doing what I asked for. So as you can see, sometimes it takes a
little bit of editing. This paused out, so
I'm going to click the continue generating button. It should just
continue generating. We don't need to,
You used to have to prompt it and tell
it to continue. But this is looking pretty good. It gives us ideas for shots. It gives us all kinds of
things that are great. I'm going to take
this a step further and show you a really cool tool, AI tool called Pictoris. Victory ai is an
automatic video creator and it can take articles or scripts and automatically create
a video for you. So I'm going to use this. I'm going to use this.
All you have to do is sign in and you can test it out. You have to pay to be able to
download your, your videos. But it's pretty
self-explanatory. You login. Again, free to sign up
just to test it out. We're going to choose
script to video because we're going
to supply a script. You can also supply,
does it link to an article which is
cool, but for this, I'm going to do the
script to Video, top three street
photography lenses. Now the important
thing is that I just need the text of the transcript. I don't need all
of the shot ideas, seen, breakdowns, etc. And so what I'm going to do is ask Fujifilm, ask Chad GBD, rewrite this, but just include the transcript
of the voice over. No quotations, no
shot ideas, etc. And hopefully it will give
me back just the text. Okay, So this is gonna be a lot cleaner script,
which is nice. Alright, so I just typed in
voice-over time calculator because what you would want to do is actually
read through this, make sure it's 60 s or less. Because sometimes it comes
back longer and I asked it to write more or
include more information. So we might have
to cut this down. So I'm just going to, let's
try the voice ram.com. And it has estimates
right here is just a basic comp calculation. Going to paste Calculate, so it does say 2 min. So I'm gonna do a little
bit of editing of this, but I'll do that in victory AIs. So I'm gonna come over
here, paste my text. Now something to note is that there's going to be new scenes every time there's a
period and a line break. And so what's going to
happen is it's going to automatically add
text to the video. And that's cool because
it's good for social media, but we don't want the text
lines to be too long. So I'm going to just
speed through this really quick and actually
edit this script. Alright, so I condense
this quite a bit. I also added some of my
own personal knowledge of camera lenses
and I swapped one out that I particularly in
particular like myself. And as you can see, it's much more condensed. I did the time estimate and
it said about a minute. So we're gonna see
how this goes. So I'm going to click Proceed. It's going to take
that information. But first we have to pick a, a template for this. So it's basically
just the text style. And we can edit this later on. But I kinda like this
flashlight one for this video, it kinda goes with my brand
and this victory tool, you can include your
own branding colors, fonts, logos that will
be added as well. Here we're gonna pick
our aspect ratio. I'm going to choose
nine by 16 for social media reals or shorts. And you can also change this
later on automatically. There's a button that
allows you to do this. What you're going to see is several of scenes
where the texts pops up and videos of I'm assuming cameras
that it's going to add, it's getting videos
from story blocks. And so that's why you
have to pay to upgrade, to use the stock footage and
use the tool to download it. But it's pretty incredible. Will be able to
swap out the video, will even be able to create
an AI based voice-over. Alright, so here we
can preview here the top street
photography lenses, so we can preview
each slide. Foreseen. It does add music
which we can change. As you can see, some
of the B-roll is good, that's behind the scenes. Some of it's not that great and won't be
able to change it. So what we're going
to do is first, we're going to edit the text. Over on the left you see all the different
things you can add it. But just to quickly show you, I'm going to move the text down. The size is actually
pretty good. It's not too big. And so I'm going to
click this button to apply a position to all scenes. Then what I'm going to
do is go to visuals. And for each scene I'm just
going to check it out. And I'm just going to search for street photography first because this isn't exactly
what I wanted. Maybe like the street
photographer is something that I would use and then I'm just going to click
it and it swaps it in place. You can easily drag and drop it, move it around, you
can preview it. Okay, So pretty cool. And this is like
super quick and easy. This isn't gonna be a Oscar worthy cinematography
masterpiece. Then I would just go through and try to find
the right B-roll. Sometimes it's good,
sometimes it's not an icon. So I'm going to click maybe
this one. There we go. And just go through it. I'm not going to waste your
time to do that though. I want to show you the audio features so we can
choose different songs, so we can choose a mood. Maybe we want something a
little bit more raw and gritty. Let's apply that. Let's find something that might fit our mood a
little bit better. Let's look for pop
rock songs genre. Alright, so we're good. You can use this one,
kick out the can. It's going to apply
it to the video. But there's another
tab for voice-over. You can also upload. So for voice-over,
what you can do is you can click to record voice, record voice-over, or you can
use one of their AI voice. Voices. Welcome to victory. It has been shown. Welcome to victory. It has been shown
that video increases. Welcome to victory. It has been shown that video. Welcome to Pictoris.
Welcome to pick. So let's just apply Jackson, and it's going to apply
it to all the videos. When you click preview, it's going to take a beat
to actually play it. For Nikon users, the Nikon 35 mm F 1.8 g is a stellar choice. Yeah, so you would want
to go in and edit that. That's why doing
your own voice-over might make more sense. But that's pretty crazy. It's not terrible. And these AI voices are
getting better and better. There's all these other
options for adding text. You can add visuals
like stickers or elements like stickers and gifts and all that kinda stuff. Down below you have
your branding. And what I'm gonna
do now is just go through edit the scenes
and finish it up. Alright, so here's the final version
that it came up with. It ended up being
about a minute and 24 s. So I could go in there, delete some stuff, maybe
make these videos, individual ones on each
brand that might even work better for my content, for keywords and everything. But let me just show you the
first few seconds the top street photography
lenses for Nikon users, the Nikon 35 mm, F 1.8 g is a stellar choice with its wide aperture and excellent
low light performance. It brings the streets to life. Other options or the
Nikon 50 million. Not bad, right? So again, you want to be as personal
and unique as possible, changing up the script. But for speeding up the process
of creating a short form, short for YouTube or a
reel for Instagram TikTok. It did a pretty good job. So now let's head
over to mid journey. Actually discord to see what
I did to create images. And I'm not going to show
you the whole process. I already did this and I
just wanted to talk through the prompts to show you
how I got my end result. I started with a prompt for
a street photographer in Bangkok, a photo
hyper-realistic. It didn't really come
back with what I wanted. So I always more specific. I changed it up. I did an experiment, an illustration of a
street photographer juggling for different cameras. I change the aspect ratio, assuming this would be four. If I put it on YouTube. This is okay, interesting, but not exactly the style. It could be used for
something but not exactly. I then did a more
specific prompt of photoshopped photo of a street photographer in Bangkok with a camera around his neck. He has two other camera
lenses in each hand, hyper-realistic, 16 by nine. It didn't follow the prompt
of two lenses in each hand, but it did a pretty good job
at a photo-realistic image. Even the cameras themselves
look pretty dang, good. I tried doing a different
prompt for digital art. I'm trying to get
this idea of like juggling ideas for cameras. This one was pretty
cool digital art of a photographer
juggling a bunch of modern cameras and lenses. I put modern in there because the other one came
back with like these retro style cameras. Camera lenses are in the air. The background is
an illustration of a cityscape trying to tie in that idea of
street photography. Then I liked that. So I did some variations. But I hit the jackpot
with my next prompt, which was a photo of
a photographer trying to decide which lends to buy. There are four lens options. He's looking confused at
each lens hyper-realistic. The four camera
brands are kin and nikon, Fujifilm and Sony. It didn't fall
follow the number. But it got the idea really well. And the quality of this photo or this
image is incredible. And it totally gets that idea. It's a captivating
image. It's interesting. I did a variation of it, and I really like this first and the fourth one
because the person, a high person is
looking at the camera. So that's pretty,
pretty dang good. So overall, I think these
images are great to use. I would throw them in Canva, add some texts for the title
of my video on top of it. Use Photoshop or
whatever tool you use. But pretty dang, good
results from mid journey. Just another step,
saving me lots of time. I don't have this many lenses. I don't have this background. It's not mean necessarily,
but that's okay. It wasn't I wasn't
trying to make it look like me for this example, but still a pretty
captivating image. Let's go over to YouTube
and I'm going to show you how to fix captions. Suppose we upload that
video to YouTube. We can edit captions
with chat GPT. I'm on the backend of
my YouTube Studio. I'm going to find one
of these other sort of shorter videos. And let's go to the caption. So what should my
next Nikon lens B? That's a great one. So let's look at subtitles. Here are the automatic ones. Whenever you add a YouTube video and publish it, transcribes it. But as you can see here, it's just a big block of texts. There's no capitalization,
no punctuation. It, it really doesn't
make much sense. So let's head over
to you to chat GBT. I'm going to give it a role. You are now a closed
caption editor. Fix this text. I just plugged in
this big block of text and came back with, this is a good lens question
from the community. It has periods, it has commas, proper punctuation,
proper capitalization. It's literally doing work
that would take me ten, 15 min to edit or someone
else and doing it in the background within
thirty-seconds. Now I can go back to
my YouTube video. I can paste this text
and click Edit timings. And it's automatically going
to time it out for us. Now we can play this. This is a good lens question
from the community. Okay, So on dry, Andre, as they have the Nikon D
7,500, they've got the 35. What's even crazier is
it takes out some of the repetitions and words that I repeat and then you
just click publish it. So I just wanted to show you another case study
or use of chat GPT. So this has been a
long case study, but I hope it shows
you from beginning to end how I've saved literally hours and hours of time using these AI tools that you've
learned so far in this class. Let me know if you want to
see any more real live demos. If you have questions about
this, about my process. And I hope that this has
helped you Have a great day.
12. Conclusion: Thank you so much for
watching this course. I know it was a super quick
one and that was my goal with this class to help you get started with these
tools quickly. They're not that
difficult to use once you understand
the basic process. And I think it's best to
just play around with it. Get inspiration from what
other people are doing. So go have fun with these tools. If you have questions, if you need additional help, post a question, I'm
here to help you out. Otherwise, I hope to see you in another class and as always, have a beautiful day. Bye
Phil Ebiner, Video | Photo | Design
