Text separation and weights to drive image creation - Midjourney Tutorial

- [Instructor] In this video, we're going to look at how we can add parameters to our text prompt in order to help guide Midjourney to get the type of artwork that we're looking for. We're going to look at this in two ways. One way is to essentially separate words when necessary. The other way is to give specific words a higher weight than others so the balance of that word has a stronger influence than all of the other words in a prompt. So how can this be done? First of all, let me write a prompt here using version five. You can follow along with this version or whatever version's at your disposal. I want to press the forward slash key followed by imagine prompt, and I'm going to type in a field filled with hot dogs. I'll follow that with a comma. And let's play around with the aspect ratio by going --ar. We'll make sure that it's set to 16:9, and I'll press the return key, and let's wait to see what it creates. So here is the end result. We can see clearly, this is a field filled with different styles of hot dogs, some with buns. But let's say that what we were actually looking for is instead of the hot dog, we eat, just dogs that are hot. What I'm going to do is take this prompt, I'm going to copy it by selecting it, and then pressing command C or control C, I'll go to the message area and then enter that imagine prompt, and let's paste the prompt and make one slight change, and that is between the words hot and dogs, I'm going to enter colon, colon. What this does is tells Midjourney to separate these two words, and in this instance, it's going to interpret the text prompt much differently. As you can see, this is a different interpretation. Dogs that are panting in some cases because they're hot versus the actual hot dog that we eat shown in the other results. Now, in addition to this, we can actually give a stronger weight to specific words so that it has more of an influence on the end result. So let's enter the following prompts. I'm going to use that imagine prompt like we've done before. I'm going to enter a neon dream filled with creativity. I'm going to enter a colon and give this a number, which is going to be a higher weight or influence over other words in the text prompt. I'm going to give this the number two, and I'm going to follow this and magic, which I'm going to give also a high weight of the value of five, and that's the maximum. Let's follow this with --ar. We'll make this an 8x10 image, and knowing that I in this instance am using V5. Now, one other cool thing once you're on a paid version is that we could start multiple prompts being generated at once. In fact, I'm going to just actually select all the text from the prompt that I just entered and enter some different weight values for words within this prompt. So I'm going to type in imagine prompt again, just paste everything here. And no need for V5, 'cause that's just going to be appended at the end, but this time I'm going to give a higher weight to creativity, a lower numbered weight to magic, so creativity is five, magic is one, and I'm going to add a high weight also to neon. Okay, let's press return. Now we have the first job already started and the second job about to start. Okay, so here's the result of the first prompt where creativity and magic had a higher word weight than everything else inside the prompts. So we could see here it's interpretation. I could see here some more magical elements in a few of these images. And let's look at that second prompt where neon has a very high weight, and creativity and magic having a lower weight. So I see much more neon colors in some of the images that were produced here. Not to mention some symbols in here that might display creativity over magic. So just keep that in mind of how we can use separation of words, not to mention word weights, to help drive the influence of the art that is created from our text prompt.