Unilingo Founder and CEO, Farbod Mansorian, joins SlatorPod to talk about launching the leading dubbing provider for Youtube creators, enabling them to get billions of additional views.
As a regular consumer of YouTube, Farbod came across TED Talks, and after translating a talk for his mother, the idea to bring this to other non-English speaking communities was born. This idea evolved into Unilingo, where Farbod formed a team to work on dubbing YouTube videos so content creators can reach fans across the world.
Farbod reveals how he develops the business by connecting with popular creators in person at YouTube conferences around the world and scaling through word of mouth.
Asked about AI dubbing Farbod emphasizes that AI has not yet been able to capture emotions and intonations effectively, which is evidenced by lower average viewing retention rates than human-dubbed content.
Subscribe on Youtube, Apple Podcasts, Spotify, Google Podcasts, and elsewhere
Farbod discusses the pros and cons of YouTube’s multitrack audio feature, with multiple language options increasing engagement and showing dedication to the audience. However, text on screen, thumbnails, and comments pose challenges as they are not currently translated into multiple languages.
The CEO addresses the challenges faced by voice actors and how Unilingo pays their talent within hours instead of weeks. The pod rounds off with Unilingo’s roadmap, where they will continue their human-first AI-assisted approach to developing software tools that enable more voice actors to enter the dubbing space.
Transcript
Florian: Tell us a bit more about background, entrepreneurial journey, how you got started with Unilingo?
Farbod: I moved to Canada at the age of 17. My dad flew me over and he said: “You stay here for two weeks and manage everything, and I’ll be back”. Two weeks turned into two years. I had to figure out everything and anything, and that built my character and that gave me my work ethic. Then I got a degree in mechanical engineering, started multiple businesses. I started three businesses, sold two of them, and I was thinking about the next path. And growing up in Iran, I had always watched dubbed content. In fact, part of my childhood, I was thinking last night, was shaped by dubbed Bollywood movies. Shout out to Amit Pachan from the movie Bagbon. Iranians love dubbed Bollywood movies. In fact, 80% of the content we watch in Iran is imported through translation and dubbing. So I always knew that this is a thing that I want to work on and I was on a hunt. It felt like having a superpower, having access to world’s best content, speaking English, and I wanted to bring that to others and open up access to good content.
Florian: Elaborate on that example briefly, so Bollywood content dubbed into Farsi or? That’s the one language people would dub in Iran or?
Farbod: Yes. It’s the primary language we speak. There are dialects, there are different dialects. But for the most part, it’s Farsi. And there was an era of about 20 years where every Indian movie seemed to be dubbed for us. Not subtitled, but dubbed. And I think the Iranian government prefers dubbing over subtitles because they can then censor whatever they don’t want the viewer to get out or hear from the content. So, the state controls the media and there are actually a lot of comparison videos where the original character said something, set some profanity, and then in the translation that was cut out, or they always cut out if there’s any skin shown by women, they cut that out. So, yeah, just lots and lots of dubbed content that was converted for us to watch.
Florian: You had that idea that this could potentially be a business, it seems like for some time. So how did you put it into practice with Unilingo? Where did you start a business and how? Were you solo, did you team up with somebody?
Farbod: I was an avid consumer of this platform called YouTube and there is this channel called Ted Talks and Ted Conferences. These are speeches given by extremely inspiring people, a lot of whom have lived through difficult times and have survived through the pain and are now telling their stories. And I was drawn to their stories and listening to a lot of them and watching, they’re highly visual. When I just exited previous company that I had, I had a conversation with my mother and she was finding it difficult to get going in life. She just didn’t have motivation. So I came across a Ted Talk called “In the Mind of a Master Procrastinator” by Tim Urban and found it to be both entertaining, humorous and educational. And I translated it for her and she liked it. And so I said, well, if she liked it, I can bring into others and I’m an entrepreneur. And so, I looked at what is the up-and-coming community on YouTube and one of them turned out to be Latin America. And also in the US, 40 million people speak Spanish as a first language. So I dubbed the video and put it on YouTube and wasn’t thinking about exactly where I wanted the business to go, but I just wanted to release it and left my phone and came back to it a week later and the video had 2.5 million views and that gave me the start. So I’d contracted this person to do the dubs, but I didn’t know how it worked. So, just like how I do most things, I packed up and moved down to Argentina for three months and spent tens of hours, ended up becoming hundreds of hours working with directors, voice actors, translators, audio engineers. And what I quickly realized is they’re just using inefficient and Asian technology and that they’re going to get disrupted one day. But just knowing that, I came back and signed new creators. Well, it was Ted Talk, but then I focused on creators, creator generated content, and we signed our first creator – Derek runs a YouTube channel called Veritasium. And now that channel is at 2 million subscribers. And he was able to unlock hundreds of thousands of dollars from just converting his existing back catalog.
Florian: As the kind of very mainstream YouTube maybe like 40+ consumer like I am. Initially, you’re not really aware that there’s all these hundreds and hundreds of maybe even thousands by now, hundreds of thousands of YouTube creators, right? You watch your favorite show, whatever else on YouTube. Were you in this kind of YouTube creator environment before or was it kind of converging along with the dubbing idea that you became more familiar with that?
Farbod: It converged along with the dubbing idea. I was not in the creator economy at all. You know, it’s like, my approach is to show up. I may be a little old school that way, but I just show. So, there’s a YouTube conference called VidCon with tens of thousands of attendees and it happens in Los Angeles every year. So I found out that they have a volunteer program briefly. I don’t think they do it anymore. And I showed up, signed up, and when I got to the city, I printed a shirt that said, let me translate your videos in other languages. And I was wearing that T-shirt and this was years ago. So, I realized that the creators are showing up a day early at a booth to pick up their badges. So I just stood by the booth and looked at all the faces and found one face that I could recognize and that was Derek. So I went up to him and I said, hey, let me translate your videos. It says it right here if you can’t tell. And he said, okay. So I put a team together and just kept at it and did not give up. At the beginning, there was a lot of ridicule. A lot of people sort of laughed at me and said, dubbing YouTube videos? That sounds pretty peculiar. But if you tell me don’t do that, I will do that. So that’s what I did.
Florian: That’s the simplest possible elevator pitch I was going to ask you about, well, how do you now go about pitching creators? Like what’s the stats, the metrics you use? Because it’s probably gotten a little more sophisticated since the T-shirt – “let me translate your video” days. So if you bump into a big creator today, virtually or maybe in person at a conference, what’s that 20-30 second metric and elevator pitch you give them for getting their videos translated into a different language?
Farbod: It depends on their size, but if they’re a bigger creator, I approach it from a bit of an emotional angle. I say, look, you have billions of views. You almost owe it to your audience to translate your content. Effectively, you’re leaving money on the table by not reaching fans regardless of their geography and language. And what we’ve seen from having now run over 100 YouTube channels and working with many creators is every time you go into a new language, you unlock about 10% of additional views compared to your total lifetime viewership. So if you have a billion views and you go into Spanish, there’s a high likelihood that you will unlock another 100 million views by going into that language. So today, Unilingo has generated 10.1 billion views across ten languages on YouTube, $11 million in additional ad revenue, and 55 million subscribers by just taking existing back catalog and future back catalog and converting it into other languages. We’ve never sent a cold email at YouTube. I think when you send emails to creators, the emails get lost in a vast sea of their inbox. So I’ve showed up to every single YouTube-related conference in the past since I started the company in Singapore, London, Dubai, Mexico, LA. So that’s how I’ve got the first few. But then from there on, it’s just word-of-mouth – creators tell other creators to do it. But we also know that there is a pricing elasticity. We know that the desire to translate on the part of the creator is pretty high. And for the top creators, translation dollars is not a deciding factor, but for the long tail, it is. So this is why we have a new AI dubbing company coming on the block almost certainly every day. But I think the mistake they’re making is they’re skipping humans and going direct to AI. Unilingo is very laser focused on retention. The average viewing retention on dubbed content at Unilingo is 51%, which is as good as the original. Whereas we’ve done A/B test, when you compare that to AI dubs, when the original character voice is synthesized, the retention is close to like 19%. And on YouTube, if you have low retention, you don’t get impressions. So it’s all about YouTube funneling a bunch of viewers to you, your video holds up, and YouTube realizes that you need to be served more impressions.
Slator 2023 Language Industry Market Report
Florian: You said 50 versus kind of 19 and I watched a bunch of the videos that you dubbed, so it’s great. It’s really well done, very professional, the voices are great, very kind of lively, obviously real voices. And so you’re saying there’s a huge difference between the retention there and then if you have what people still can tell are AI voices even now. But, I mean, obviously it’s gotten a lot better maybe over just the last six months. I mean, what are your thoughts on progress there? Where do we get there? We recently had voices on the podcast. They help with gaming voices, making them more natural. And I mean, sometimes it’s very hard for me to tell the difference between a natural human voice. Where are the complexities that, let’s say in 2024, as the technology even improves even further, that an actual human voice actor would still have an edge and that word would filter down to the metrics that are so crucial that it’s just outlined? I mean, retention on YouTube, that’s what you care about if you’re a creator.
Farbod: If you fast forward to 100 years from now, it’s more likely than not that most videos are consumed in a completely different way and probably fully AI dubbed with very minimal work on the part of humans. But my estimate is that if you put a gun on my head and say – “how many years before top creators start considering AI dubs?” – I would say about somewhere between three to five years. With gaming, there’s a trick. If you use synthetic voice and then you work on the speech synthesis in post-production, you can make it sound normal, closer to normal. But one thing that no AI dubbing company or rather any company has figured out is emotions and intonations. We dub Will Smith’s video for a YouTube channel and the clip was from “Prince of Bel-Air”. And there’s a scene where Will Smith gets really emotional, gets really sad, and he’s confronting his dad and I remember the voice actor in the studio stepped out. When we got to that scene, he said, all right, I need to step out of this recording studio. So he stepped out, he was fully happy when he stepped out. When he came back in, this man had transformed into just a sad human being. His eyes were red and he was on the verge of crying. And I said, well, what did you do? He said, I thought of a sad story. And then he said, hit record button now. So the audio engineer started recording him and when you now listen back to that piece of dub, it is beautifully done. It turns out that there aren’t three variations of happy, sad, anger. There are a thousand variations. And on YouTube, if you look at good creators, the first 10 seconds and how they deliver it is really important. Say, I am going to do this, or when they get angry, they can be a little angry, they can be really angry and if they’re sad, they can be a little sad, a lot sad. So I think emotions is something that still very important that nobody has addressed.
Florian: Is it really that you said the first 10 seconds, how important is it that really those first 10, 15, 20 seconds of the video are just hitting like a home run? Is it that crucial or is YouTube kind of the algo that dumb? That’s it, you just get the first 20, 30 seconds right and then you’re set.
Farbod: No, the algorithm is actually pretty smart. There are 500 to 1000 hours of content being uploaded to YouTube every minute. The sheer amount of content being uploaded is a lot. So YouTube serves a piece of content to someone to decide whether this is good or not. Can they increase user watch time on the platform? YouTube changed their North Star from views to user watch time many years ago. So the longer you can keep a user on the platform, the more you get rewarded. And therefore those first 10 seconds are extremely crucial, extremely crucial. Why? Because people click on a video because they were served video on the side of another video as part of suggested videos or on homepage or browse and then they click on it. Then within about 5 seconds they decide whether they want to keep watching or click away. So, typically, when you look at the retention graph of most videos, there is a 60% drop in the first 20 seconds. So you got to hook them right there and then and you got to rehook them every 20 seconds, every 30 seconds. So you could say, well, we are going to use human voice actors for the first 20 seconds and then somewhere along the way switch over to AI and say, well the moment the viewer realizes it’s AI, they’re going to click away.
Florian: Let’s talk about multitrack audio on YouTube. A feature that they rolled out or have been rolling out for seemingly now two years, which seems like a huge deal if you consider what you just mentioned with the stats. So previously, before multitrack audio you actually had to upload different videos, right? A new video that was dubbed. Now certain creators have the option of clicking like with Netflix, right? You get ten languages. So tell us more about this feature. Why is it rolled out so slowly? How is it important for view count and how have creators taken to it so far?
Farbod: I was aware of the multi-language feature or the ability to switch between different languages since 2013 on Netflix and it was just a matter of time before any of these platforms copy and pasted that feature. I just didn’t know when. And YouTube is a great leader in video distribution so they were of course the first to do it and, actually, they introduced the multi-language feature back in 2019 on Netflix movies that were hosted on YouTube but they were not getting a whole lot of traction. But if your question is what are the advantages and disadvantages of multi-language feature versus starting a different local channel?
Florian: Let’s talk about that first.
Farbod: What is Reddit known for? Community, a sense of community. YouTube to a large degree, if you put shorts aside, if you look at long, is also known for belonging, being part of a fan base for a creator. And starting a separate YouTube channel where the thumbnail, the banner, the video description, the playlist, the about page, the comments, everything is in your native language, it really increases engagement. It really gives people a sense that okay, this creator went extra the extra mile to localize all the metadata for me. So that’s the advantage of starting a localized channel. The disadvantage of starting it is… We were the first company to upload the multi-audio track for the very first creator using it at scale, which is MrBeast. And we worked directly with the product manager at YouTube on this feature and since then the feature has come a long way, which is great to see. But Jimmy said to me, I would rather have 100 million views on one piece of content on my main channel than have 10 million views on ten localized channels. And that makes sense. The higher the views, the better it looks for brand deals that he could get just a better perception, but also it’s just better. But the thing is there are three disadvantages when you look at the multilanguage feature. Number one, if there’s text on screen, unless YouTube retrofits their engineering infrastructure, which would take many quarters to allow creators to upload multiple videos with multiple text on screen in different languages, that problem is not going to be solved. I hope that YouTube works on it. Then the second one is in the same sphere as the first problem, which is thumbnails. If a creator uses thumbnails in a video then they can’t upload multiple thumbnails. And the third and most important one, which is kind of going back to the reddit analogy is comments. If you go to the last MrBeast video and sort views by newest, you will see that the comments are written in different languages and you often find people saying if there is anybody else speaking Russian here, please like this. There are so many different languages, there are twelve languages on most of his videos now. And so people just sort of lose a sense of community.
Florian: It would be interesting. Think about it. I’m sure this is a pretty big engineering challenge for YouTube but if you landed on a channel and then you could click, okay, I want to have this channel in German or Spanish or French or maybe if the original is English and then instantly the interface changes but the view counts maybe remain right? Because if that’s so critical to the creators. Yeah, those are all great points because I was kind of more single-mindedly, convenient, great. I can now watch it in a different language. But all the other components, what we in the industry would call like the localization of the whole interface and everything would just be I mean it would still be in English and all the metadata and all the descriptions and yeah, the thumbnails, etc.
Slator Pro Guide: Finding Growth in Adjacent Services
Farbod: To a large degree, YouTube does allow you to localize the about page in the backend. They also allow you to localize the titles and the video description in the backend, there’s an interface, you can type it. It’s more like the hard assets that are burnt in the thumbnails, the banners as well as the comments. But I truly hope that this all works out so that we don’t have to launch more localized channels. It’s more difficult. This would make it easier to just upload different dubbed audios.
Florian: You guys also manage those localized channels so that’s a bit of a service that you do? So if you’re a creator like – look man, I’m busy, I don’t want to deal with all this metadata stuff, and managing these channels – is that something you do as well?
Farbod: We’re now doing it at scale, so we use software to manage channels. We don’t grow one to one ratio for every channel, we have one channel manager. It’s more of a scalable approach but yes, what we’ve realized the success that Unilingo has had is creators really don’t want to do anything other than be shopping from the camera and make content. They want to outsource everything else: editing, design, especially translation. Like if I’m a creator and I have made 500 pieces of content the first question is well, what videos do I translate for what language? The localization is like a Lego and there is a translator who is the one piece and then there is the dubber, there is the director and then there is the posting part of it. So Unilingo is working on a single system of record to effectively allow creators to translate, manage and distribute their content for global fan base so creators don’t have to do anything.
Florian: Let’s go back to the multi-track audio thing. So why does somebody like MrBeast have it now but PewDiePie millions of subscribers as well doesn’t have it? So what’s the kind of magic here that YouTube allows some creators to have it and others don’t?
Farbod: It’s a decision made by YouTube and their team and how they plan to release it to everyone. But I think, effectively, what they don’t want to do is there are two things. They don’t want creators to regret using this feature, so they want to minimize unregretted spent resources on this feature and then that will look bad on the side of YouTube. My hunch is they mostly enable this feature for creators that are more likely to have success with the multi-language feature. And then there’s another side which is the reason why YouTube does not open this up to every creator is a trust and safety issue. The first five dubbed pieces of content that we did for MrBeast had to be submitted to the trust and safety team at Google for them to review the translation. If you open this feature up to every creator and a creator, let’s say, has an original video that talks about birds and flowers, but then the audio feature in Spanish is something extremely bad. YouTube would have to 10x its engineering resources to use speech recognition, figure out what was in that audio, and to effectively maybe bring the audio down. So it’s also a matter of trust and safety. If you put profanity in your audio and the video is about something else, YouTube will get hurt.
Florian: There’s a lot of stuff we could think about here. Yeah, there’s infinite trolling potential. Now, you are very deliberate in hiring, retaining really well-known dubbing talent. So you just launched a podcast as well called The Unified Podcast that came about during research for this podcast. And you’re starting to interview, like, the voice of Spiderman, the voice of Luke Skywalker. Goku. How do you connect with these top voice actors? How do you retain them? Is that a big challenge to get them to be available to you?
Farbod: I entered this space as a viewer, and if you enter this localization space as a viewer, you know how important voice actors are. And you know that voice actors are effectively the celebrities of their own countries, and they’re treated like celebrities. So when we interviewed Beto Castillo or Mario Castañeda – the top two most well-known voice actors, these folks are celebrities, and we should talk to them because, they have a lot of insight, but doing a podcast with them will probably be a good idea. But the way we get them is, again, we started a studio in Mexico City, and so I’m very well connected to the dubbing industry, and I just asked someone, hey, do you know so and so? And they connect me. But the way we retain them is slightly different. The localization monopolies right now, in my opinion, are pushing out voice actors and translators by using aging and inefficient technology, making their lives extremely difficult, making them to go to recording booths as well as underpaying them and paying them with lots of delays. So localization monopolies generated $3.2 billion last year. Of that, probably less than 5% went to the actual folks who made the work happen, like the translators and the voice actors. So what we do, we bring that up, we pay those voice actors more and the translators more. We are also building tools to make their lives easier, but we also pay them fast. Like, the average voice actor takes 45 days to see a check. We strive to pay them within 48 hours.
Florian: This is a common theme, though, in the language industry. Just these, really lengthy payment terms. I never quite understood those. It’s kind of a bit of CFO type of engineering in the back.
Farbod: I spent close to 1000 hours in dubbing studios in three countries and these are top dubbing studios. They do Hollywood and Disney and Netflix and HBO. I know exactly what they’re doing. And this is why I’m excited about Unilingo. Because we’re effectively going to become the hero of voice actors. We are not telling voice actors you’re going to be replaced by AI. Voice actors are still scheduled via WhatsApp? For God’s sake? Use something else.
Florian: Are you competing with some of these big dubbing providers, they work for Hollywood, right? For the big entertainment, 2 hours, whatever, cinematic releases. So you’re competing in a sense for the same elite voice actor pool, when you’re dubbing for the big YouTube creators than these big dubbing studios, for the Hollywood studios. You’re competing for the same talent pool?
Farbod: Yes, the talent pool is small. Why? Because the existing talent pool is leaving the industry. They’re generally not happy. If you watch one of our podcasts, the first question I asked to Beto is what is one thing you want to change about the industry? What does he say? He says I want to get paid more. Well, A – we pay the voice actors more. We treat them with respect, with dignity, fairness, and they see results of their work. If you’re a voice actor and work on a show, you never see the result of your work. You don’t get credit for as much as you deserve. So we are competing, but we are also building tools to allow new voice actors to enter this space.
Florian: How much potential do you think there still is in this creator market? Are we just barely scratching the surface? And if YouTube decides to go much broader with this audio track feature, it’s just going to explode? Or do you think it’s more of an incremental thing, like it’s been maybe over the past two years? Where do you see this going?
Farbod: I think more and more people are going to create content and more and more people are going to watch it at a global scale. So what that means is language is going to be an integral part of consuming content. And it just so happens that right now, 80% of world’s top-performing videos are English based. But 60% of the world population does not speak English as a first, second or third language. My assumption is well, not an assumption, but it’s largely due to the fact that the large distribution platforms were founded in English-speaking countries. So therefore a lot of the content that is created is in English. So I think this is going to explode. I think the multi-language feature is going to be copied and pasted on other platforms. I think it’s going to be on TikTok. I think it’s going to be on IG reels. And what we know is the demand is high on the side of localization, but because these localization companies have aging and efficient technology, they can’t keep up. Whereas for us it’s much easier. We are much more nimble. I approach the dubbing industry from a completely different standpoint. I’m a mechanical engineer, so I use a lot of friends’ principled thinking. And one of the questions I kept asking back in 2018 when I first got into dubbing is why does a voice actor have to drive to a recording booth to record a 30-minute session and then drive back? Why don’t we do it remotely? So in 2018 I hired a voice actor in Spain and had him be recorded in the US. And there was a lot of resistance. But between 2020 and 2022 we saw a new movement and remote recording became a norm.
Slator Machine Translation Expert-in-the-Loop Report
Florian: Have you looked at all in the Asian part of the market – China, Southeast Asia? I think. Southeast Asia YouTube might be very big, but in China it’s blocked, right? You can’t watch it. Have you done any business in that area or is that completely closed off?
Farbod: There are so many Indonesian-speaking people on YouTube and we have no idea how big it is. We do Indonesian, we do Thai, we do Viet. There are pretty large communities on the distribution platforms. In terms of China, China is like this 500-pound gorilla that comes up in every conversation. But the way we address it at Unilingo – we are platform-agnostic. We are not married to YouTube. We want to open up access to good content on any user-generated content platform. One of the first creators we work with was a Udemy instructor who was selling coding courses in English. And so we came along and we dubbed it at a low cost. And he started selling his coding courses in China and we dubbed it into Mandarin and Cantonese and he was able to convert another, I think, 3000 students. The thing is, all these AI dubbing companies first focus on creator-generated content, then it doesn’t perform because intonation emotion is bad. Then they have to get out of it and then focus on educational content or voices with their monotonic, not lots of head movements. Most creator content is lots of shouting, lots of characters, lots of age cases. So they focus on that. But they also fail to realize that teaching and learning is a personal experience between the student and the instructor, the professor, the teacher and you want to try to remove the human. And I don’t want to come up as an anti-AI person. Unilingo is now using AI at almost every aspect of the localization process. But we do it in a way that doesn’t affect the retention.
Florian: What you just described before, the shouting, the head. I mean if you watch MrBeast video, PewDiePie or whatever, like literally the first 1 minute it’s just full of action, emotion and I don’t really see even slightly monotone, AIish voice doing this or even one of these more advanced voices. It would be just challenging to get that across. Yeah, China, I lived there for a while, right? I don’t even know what people are using as the YouTube replacement and everything’s on WeChat, Youku was a thing.
Farbod: The equivalent for YouTube and TikTok over there. Well, they have TikTok, I guess, but they also have Baidu and BiliBili and a lot of American creator content or English content is pirated, illegally dubbed and republished to those platforms. So we are working on that as well. But there is a huge market in China and good content is heavily undersupplied in that market. It’s the same with Iran. Our access to YouTube, Instagram, Facebook is blocked. So I can’t wait for this government to go and there is a revolution and all this access opens up because about 100 million Iranians are going to come on YouTube for the first time. And that is so exciting to me and good luck trying to compete with me on that front. I’ll be the first person to dub that.
Florian: How is the VPN penetration in Iran? Because in China, you have a VPN, it works and then it occasionally kicks you out. But is it tougher in Iran or possible?
Farbod: There is a number that came out that 30 million Iranians are now using US-based VPNs. Even the government officials are using VPNs to access Instagram.
Florian: I would assume the same thing is happening in China. It was just it got harder and harder. I remember I left there in 2013 and it was okayish. I think now it’s gotten a little more tricky. Now what are you doing requires, obviously, creativity but also maybe some capital at some point because the opportunity is very large. So far I think you’ve been bootstrapping. What about partnering with one of these VC funds or other funds, pots of money that are out there?
Slator 2022 Language Industry M&A and Funding Report
Farbod: I went to a thrift store and I said give me the largest hat that you have. And so he gave me the largest hat. Now I’m going around and asking these two companies to please donate because we are about to go out of business. We don’t need them. We are fine doing what we do. I think we have a very methodical approach to growth and there will be a time where we raise, I will not say when, but look, an AI dubbing company, the founder of that company had reached out to me a year ago asking for feedback and he ended up raising $3 million seat round last March. I’m very much aware of what is happening in the market. We’re doing it a little bit more methodical without sharing more.
Florian: There’s pros and cons to all of it. I mean, taking out outside capital can be disruptive as well. Let me leave it there. Anything that’s on the roadmap for you guys this year, next year, any new product services that are in the pipeline?
Farbod: We are taking a human-first-AI-assisted approach and building a set of software tools around it to effectively allow more voice actors to enter the space so that we can dub more user-generated content. And the way we will use AI is used in such a way that does not negatively affect retention. So we don’t want creators to regret spending their hard-earned dollars on translation and have bad ROI. We have hundreds of creators on our waitlist. We have the 60% of top 1000 creators on our waitlist right now. And we could just open up the floodgate, charge them however much these dubbing companies are charging Netflix in ten languages and send them the giant invoices every month. But chances are, some of them, the views, the videos are not going to perform, and then they’re going to regret spending money on it, and then it’s going to look bad. So we take the approach is methodical, like bring the cost down, launch them into one language. If that’s successful, launch them into another and another. When Jimmy Donaldson came to me, he said, I want you to launch me in ten languages as soon as possible. He didn’t want dubbing for two years, and all of a sudden he did because I gave him a pretty good pitch. But I said, look, let’s get you in Spanish first, and if that’s successful, then Portuguese, then Russian, then Hindi, then Arabic, then French, then Germany, and the subsequent languages. So we just want to minimize unregretted, spent dollars on translation.
Florian: It’s fascinating and so different from a lot of the rest of the language industry because here we have immediate ROI figures, and even public figures at that, right? In a lot of the rest of the localization space, you’d be kind of guessing, like, okay, we launched that global marketing campaign in ten languages. Well, here’s our very complicated dashboard to assess if the ROI of this localization push was successful. And then, of course, in a lot of the rest of the language industry, it’s a pure cost. There’s not an eye, it’s just an input cost, right? Now with you and dubbing, and specifically dubbing for YouTube critics, it’s immediate, it’s public, and it’s a very simple equation. I spent X for this and I got Y out of it. And should I do it or not in the future? So I understand why you want to be careful because otherwise it might be damaging. But, fascinating. So public, very different from the rest of the industry.
Farbod: This is why voice actors really love to work with Unilingo. They immediately see the outcome of their work through comments and through views.
Your information will not be shared with third parties. No Spam.