Ep 31 – Using A.I. to Analyze a screenplay

GEOFFREY D. CALHOUN: All right. Welcome to the successful screenwriter podcast where we discuss anything and everything screenwriting here, we interview successful screenwriters and filmmakers to find out just what it takes to make it in the industry. All right, welcome to the podcast. Yes, this is an awesome guest we have on today Nadira Azar. Now she is the founder of this really cool company called script book that has used A I technology to analyze screenplays. Nadira. Thank you for being on with us.

NADIRA AZAR: Thank you for the invite. All right.

GEOFFREY D. CALHOUN: So before we get started, before we roll into this, I want to first kind of get your origin story, how this all came together.

NADIRA AZAR: Of course, A script book was found five years ago. And actually the conception of the company happened years before that when I was still in universe in university, my background is engineering and economics. And when I was doing my thesis, my master’s thesis, I was looking for a subject or topic really. And, and and I was diving into a lot of research surrounding predictive modeling. And I came across a lot of scientific research in the film space and that kind of struck a, a, a nerve with me for some reason I was very interested in, in in this industry. So, and that’s where I started to do all of the research, seeing what kind of technology was out there for specifically for, for the film industry. And I wrote my dissertation on the topic predictive modeling, specifically for film. And you know, and then the thesis ended up somewhere in a drawer for years, like any thesis. Right. Right. And then it wasn’t until years after that, that I kind of felt like it was the right time for some reason to to kind of dig deeper, get the that dissertation out of the drawer and dust it off and see if I could do something with it. So that’s really the conception of the of the company.

GEOFFREY D. CALHOUN: So when bringing it all together, did you bring in screenwriters to kind of help create this this A I?

NADIRA AZAR: Well, when we, I have to say when, when I started the company, I honestly thought building the technology would be somewhat feasible. But early in the process, we kind of found out that it was al almost impossible to build a script analysis system that is fully automated. So, and I remember when we were hitting the, the phase in which we were annotating, labeling data. For example, I’ll just give you an example on our the platform. If you, if you look at, at the platform, we have a few features like liability and the protagonist antagonist. Now stuff like that. Are you can’t really have them automated, you know, you can’t really develop an A I model that does that from scratch. You need to, you need to label data and, and what we did, we spent the whole summer bringing and film students. And we wanted them to kind of label scripts, you know, just read scripts and telling us who is the protagonist antagonist also sign likeability scores using certain scales. So all of that label data, we used to develop an A I model. Oh, fascinating.

GEOFFREY D. CALHOUN: I stumbled upon your, your, your service because it, it was, it, it came out kind of with a bang a few years ago. It was everywhere. And then the other day, I was like, I’m gonna look into this. So I I, you have a a free option where you can go on there, you can upload your script and then you can have it analyzed. So I was like, OK, I’m gonna throw a couple of scripts at this software, see what it’s got. And I was surprised it was really interesting that the, that the process does a scene analysis, it does the likability. So, I mean, the characters that I thought were likable, of course, were, of course, in a screenwriting sense, we don’t feel you have to have a likable character as long as they’re understood and, and the audience can relate to them. But I can see where the likability factor comes in. I really liked the emotion by scene. I thought that was fascinating because I think that’s something that a lot of screenwriters could actually look at their script, they could get it analyzed and then they could see a waveform and see the emotions of the screen, which is really fascinating. And then you had the p the the Bechdel test, which I think you guys call it the gender, the gender equality test. And so what that is, if anybody doesn’t know what it is, it’s do you have two named female characters that are actually prominent in the story? Are they having a conversation and is a conversation not about a man? Now, I know that sounds ridiculous. We have to have a test like that. But you would be surprised of how many scripts writers don’t realize are ignoring that part of the script. And so I love the fact that you had that and it gives you a score on, on how you do. And and I put a script in there that, that I, I kind of thought wouldn’t do well and it definitely did not do well. And then I put another script in there like that would, would do good. And it, and it did, of course. So I was honestly, I was testing your software to see how good it is. And, and so far I’ve been impressed. So the really cool thing you have is the marketplace. So tell me about that. What, what is this marketplace with the scripts?

NADIRA AZAR: Well, when to be honest, the software was built for companies initially B to B. But we saw, we have a lot of you know, channels in which we get feedback. And a lot of screenwriters have a genuine interest in, in using technology for their purpose, you know, to, to kind of understand their material from a different perspective, not just the, the usual suspect, you know, through or anything else. And I, I was actually thinking about how do I come up with, with a pricing model because I know most, most writers are always struggling. I get a lot of emails, they’re always struggling. So we needed to find a, a pricing model that works for them. But I’ve come to the conclusion, there isn’t really a pricing model that works for writers because any price is too much is, is actually too much for a writer because I personally, I also believe that it’s not a writer should pay for a service. I think it, it should be a different model in which a writer really shouldn’t spend any money on getting information on, on what whatever the story he’s he’s written. And the marketplace, it was all to be honest, it was, it started with the whole Corona thing. I was thinking about a solidarity action because I got a lot of emails of struggling writers and they wanted, you know, some help from our technology. They couldn’t afford it. And I kept thinking, I spoke to my team, I said, listen, we need to find a way to set up a marketplace and have free analysis for for writers, this is some sort of a light point that they’ll have in this dark year. And I reached out to be honest to quite a few platforms, prominent platforms that do coverage that do script hosting and not a single one of them wanted to team up for solidarity purposes. So I thought, OK, nobody is interested in solidarity. We’ll just do it, we’ll just do it by ourselves. So, yeah. And that was the market place. We just kind of launched it. A couple of months ago, we didn’t really promote it. It’s just there. And I think there are about 400 projects on the marketplace right now. It’s growing steadily. I we don’t really promote it but anybody who finds the marketplace and sees, oh, I can actually upload my scripts for free and, and have some sort of information.

GEOFFREY D. CALHOUN: Well, that’s what I want to ask you because, because I, I see, I mean, what do I know? Right. But I see potential on this marketplace for scripts that are scoring high that perhaps producers start perusing this looking for projects. Is that something that you can see as a goal? It’s definitely something that would be interesting, of course, because we, we had a, for example, a few film funds also looking for material and they were asking if we could kind of you know, be the pipeline of successful projects, commercially viable projects.

NADIRA AZAR: But I think this year is just not a benchmark in the sense that when we, we kind of had you know, scripts that did very well commercially and, and the in the predictions and we kind of, you know, I emailed those scripts to film funds and production company, but it seems like this year, not a single company has money this year.

GEOFFREY D. CALHOUN: This year is a…

NADIRA AZAR: It’s because I spoke to three film film funds about, I think three projects that were very good in, in in their predictions. And they all said our money is tied up. We didn’t, we don’t get the insurance. So I was like, OK, 2020 is a last year. So hopefully next year, but it’s definitely something that would be interesting to kind of be this funnel that that, you know, that we kind of put a producer, writer in touch with, with a production company or a film fund and help them get, get made.

GEOFFREY D. CALHOUN: I think, I, I think, I definitely see there’s, there’s possibility for it on your end. So, so I hope that that keeps continue being developed because I’m all about finding avenues for screenwriters. Now, pricing wise, this is definitely a price for a studio for a production company or for a producers, some kind of a development. This isn’t something that a writer’s gonna shed this much cash on. However, looking at it, I was always wondering if, if there is a price model or if in the future, you’re developing something a little bit more that a writer could take advantage advantage of because I saw aspects in the paid version, which is like, I don’t need any of this other stuff, but you had an option in there about budget breakdown. And I was like that fascinated me. I would love to know how much it would cost for my script to be made. So there are little things like that that I think the screenwriter could be like, yeah, I would pay for that.

GEOFFREY D. CALHOUN: So what is what is the future of script book outside of the marketplace kind of getting that going says that’s so new. What are you looking at? Are you looking at trying to branch out in, in more Indie stuff or?

NADIRA AZAR: Well, for us, I mean, our, our main goal is content analysis really. Having a I driven content analysis, that was the goal from day one. We we wanted to to deliver, to offer a new way of assessing material, a more efficient way, a quicker, a quicker way, an unbiased way. And, and we honestly, and, and I remember when we started out, we had, we were very, also very naive in our, in our expectations. We, we were just saying, and writers will love this. This is something that will help the, the you know, the, the group that doesn’t get into, you know, that doesn’t get invited to the table. And I was always about that. I really wanted to kind of give that specific group of voice and I see the difference between, let’s say Hollywood and, and the in independent world for sure. The independent filmmakers are struggling and Hollywood is kind of going to towards a different avenue I mean, if you look at the jobless things for every studio or, or, or big production company, all they do is look for data scientists, there’s nobody looking for writers, check out the jobless things that should be that should give you insight, right, in the sense that all of the studios and large production companies are heavily investing in data science. That’s why they have job listings for 30 40 50 data scientists. So that, and, and the independent industry should kind of, you know, look at that and, and, and make put one and one together. It’s, it’s, it’s, I think it’s just time, especially with this weird year. It’s time for, for, for writers, but also independent producers to kind of find this you know, the silver lining and this all, you know, between technology and their artistic and their creativeness.

GEOFFREY D. CALHOUN: I think that people don’t like change and, and, and what, what you, what you’re presenting is a change. I mean, this is huge and other than the fact that artificial intelligence has been like the main villain in any kind of sci-fi, scary film, right? So we’ll, we’ll go outside of like the Johnny Mnemonic stuff. The, this is, this is, it’s changed, it’s cool. I bet. But I saw it and I’m always like, I’m always looking for avenues that people can find and, and, and take advantage of to help their career, however they can get that edge up. So if I can upload a script on the script book and get that objective viewpoint on it. I like that. I like that a lot because maybe there’s something there that I miss, like, maybe I didn’t realize that I failed the Bechdel test. So I think that’s great and the fact that you have a free package is great. I, I think people have to get used to it and that’s tough. The, the one thing I, I would ask you about is what do you think? Because you’re an A I person, you’ve developed this, you’re, you’re, you’re an analyst, I’m talking to the person. So, what do you think about A, I created content where these people are forcing A I, they’re programming A I technology to be able to write scripts on their own, which I’ve read. Now, some of these are just memes and, and they’re jokes where they’ll say I made, an, a, I watch 1000 hours of Christmas films and it’s a joke. They’re being funny but some of them are real. They’re actually programming these scripts, or these A I technologies to write scripts and I’ve read them and they’re gibberish now, but that doesn’t mean they’re gonna be gibberish in 20 years.

GEOFFREY D. CALHOUN: Yeah.

NADIRA AZAR: Well, first of all, I think there is the first issue that I’m kind of, I worry about is the fact that writers, independent producers, let’s just call them to create a crowd that are kind of somewhat allergic to change, allergic to, you know, the fast paced environment that technology is, I worry about them because it takes them too long to get used to to innovation, to, to novelty. You know, we have been around for five years with, with our content analysis, script analysis platform. And up to until today, you still have a large group of people trying to fight a fight this because they don’t want this in this space. But on the other hand, what I worry about is that there is already new technology being launched, which like, for example, for a script, we develop deep story, which is also our own script.

NADIRA AZAR: This is something that we started out three years and a half because I knew early on that a generative A I and A I that would be able to generate content, you know, was going to you know, be, be in the world, you can’t really stop progress.

GEOFFREY D. CALHOUN: It’s gonna happen, it’s gonna happen.

NADIRA AZAR: And three years ago, you’re right, scripts that were generated by an A I were kind of, they didn’t make sense, there was no context. You know, it’s just like a a loony bin.

GEOFFREY D. CALHOUN: Like, I mean, that’s how I read it. Yeah, exactly.

NADIRA AZAR: It’s, it’s just so weird and, but that’s the thing, it was very inconsistent three years ago. But today even just with our own. A I, for example, with deep story, we, it’s also the beta, the alpha version is, is also live so people can kind of test it. If you look at the quality of what’s being generated today, it’s a massive improvement compared to three years ago. But let’s not forget that also open A I which is, you know, the big company that is being founded by Elon Musk and, and Microsoft has put a billion over a billion dollars into open A I. They trained the largest language model in the world. The largest language model. I can’t tell you how big that language model is, but it’s the best language model in the world. It’s called GP T three. And if you do some research on it, you will be shocked. I have to say what is being generated by that language model is honestly better than humans that sounds painful because in the sense, you know, writers and the creative crowd are, are only getting used to an A I that’s able to, to assess their creativeness that is able to assess their creative works. While on the other hand, the world is already launching a different A I that actually cores creates something that might, might be superior to what humans can do.

GEOFFREY D. CALHOUN: Well, in the, in the screenwriting world, it sounds apocalyptic.

NADIRA AZAR: It’s it’s a lot I have to say because III I, I’m also one of the beta users. You know, for, for the open A I model model, I have to say our own model is also very good. So I’m, I’m quite we’re competitors. Let’s just keep it that way with open A I. But I have to say when I see what it can create, I have my moments where I think, oh boy, there’s no hope for you.

GEOFFREY D. CALHOUN: Well, what kind of advice would you have for the screenwriter? Sitting at home that is, is working is trying to make this happen. Hearing from hearing that eventually one day studios will be using programs to create their own, create their, their own stories. What, what, what kind of advice would you have?

NADIRA AZAR: I think the first thing anyone in this industry starting out or being, you know, an, an old rot in this, in this business, I think it’s just important to kind of hit the reset button and just forget what you learned. It sounds very harsh but I mean, there are systems out there that will do that will write better than you do. They will do it efficiently, they will do it quicker, they never get tired, they never have a writer’s blog, they never sleep so they can start over and over again. If you’re not happy with the scene, if you’re not happy with certain lines, it never gets tired. How do, how do you compete with something like that. And I think it’s not about competition. I think, I think we, we really need to stop trying to put the A I versus humans. This is not the battler and I think humans and A I should kind of, you know, it’s like like the movies where you think you’re enemies but actually you’re fighting the same cause I think we need this, this, these two things, you know, the humans and the machines need to come together and collaborate. I think that will be the best thing that, that one can do.

GEOFFREY D. CALHOUN: So the utopian version of this, if we don’t go full terminator where we have screenwriters fighting A is for Skynet. If we go with the utopian version of this, you have screenwriters that have worked on the craft filmmakers who have worked on the craft, master the craft and they’re using the A I to enhance their own performance. Would you say that would be ultimately the most beneficial goal for this type of software?

NADIRA AZAR: I think personally that the perfect use case that I see, for example, for our own deep story. I think the perfect use case is where a writer starts writing. You know, they start writing a certain scenes and then they ask for help whenever they don’t know where the story is headed or what’s the next scene or I’m stuck, you have the A I give you ideas. Hit the gener cool.

GEOFFREY D. CALHOUN: I like that. I love that idea.

NADIRA AZAR: There you go. And, and you kind of work together with that. A I and the A I is just making suggestions, you know, it’s just like this, co writer, an artificial co writer that really doesn’t have any feelings. So you won’t hurt his feelings or feelings, you know, just delete if you don’t like and it will make the writing process far more efficient.

GEOFFREY D. CALHOUN: It, it sounds like a thought generator and I thought that’s a really, it’s a really interesting idea, could potentially be a nice solution for writer’s block. I, I like that. I, I like to think that would be used to call.

NADIRA AZAR: It the INS, but it’s like a thought generator.

GEOFFREY D. CALHOUN: No, I love that. That, that, that I think that could give everybody hope. That’s great. Well, this is, this is a really interesting software, so it’s called Script Book. And we’ve got the marketplace, we’ve talked about pricing, we’ve talked about the free service for, for the, the Indie industry. Where, what’s, what’s the future here? You have your deep story. So you have your auto your, your thought generator. What, where, where else are you going from here? Are you going to lean more into Indie stuff? Are you gonna keep going with the studio model? What, what is the, what is the goal here? Where’s the, where’s the projection?

NADIRA AZAR: To be honest, the bigger goal for scrapbook? And, and you know, one can dream is that we really want to build a new ecosystem. I think there is way too much fragmentation happening in our industry. You know, you have a, a zillion platforms, all launching their own little software and there’s nothing coming together. You need to consolidate every tool and, and bring them together in an ecosystem in an eco. Yeah, an ecosystem where anybody who’s serious about this craft and wants to create a future in this industry, you know, can find everything consol in a, in a consolidated space. I think that’s, that’s the goal. That’s the goal instead of having, you know, a lot of wannabes passing you know, the, the, the right, there’s.

GEOFFREY D. CALHOUN: There’s pockets of expertise and pockets of, I would say authentic. And, and then impostor almost. So you don’t, you never know who am I using? That is real, that can help me, who am I using? That might be predatory. So I agree with you. There, there is definitely needs to be some kind of a melding of the minds, I guess you could say but, oh, that’s a lofty goal and I wish you the best. I love it though. I mean, you, you’ve got the drive and I see the passion and when I see that passion, I, I can always, I can always believe that that’s somebody that can make it happen.

GEOFFREY D. CALHOUN: Well, we hope so.

NADIRA AZAR: We hope so for sure.

GEOFFREY D. CALHOUN: Nair. I wanted to really thank you for your time and thank you for coming on. This has been an an enlightening conversation coming into this. I knew it was gonna be, there’s gonna be some, some big thought bombs coming at me about a I because I was like, this stuff is gonna be huge eventually. So I wanted to thank you for that.

NADIRA AZAR: Thank you. It was very interesting, you know, to kind of hear your thoughts as well. So it was great speaking.


Leave a Comment