You can also find this content on ESTsoft’s official LinkedIn and Facebook.
ESTsoft, a company specializing in AI services focuses on making people's work and daily lives more convenient through AI. STUDIO PERSO, ESTsoft's global SaaS service, creates videos of an AI avatar speaking on the user's behalf. With nothing more than a script, you can make professional-quality AI avatar videos without a camera, studio, or professional crew. It helps companies produce content more efficiently when they need to make branded videos repeatedly for corporate training, lectures, internal announcements, and more.
For this edition of ITself, we spoke with STUDIO PERSO team backend developers Baek Sang-jin and Jang Hee-ju who are developing the service. We heard about the challenges they are tackling to make video creation easier and more convenient for users, and the stories behind turning AI technology into a real-world service.
Q. Hello! Could you each briefly introduce yourselves and tell us what you do?
Sang-jin) Hello. I handle backend development on the STUDIO PERSO team, mainly designing and implementing APIs. Lately, I've been especially interested in using AI to make development work more efficient, so I'm trying out various approaches.
Hee-ju) Nice to meet you! I also work in backend development on the STUDIO PERSO team. I make sure the service works properly behind the screens users see. These days, I mainly work on improving features so users can use STUDIO PERSO faster and more easily.
Q. What kind of service is STUDIO PERSO?
Sang-jin) Simply put, it's a service that creates videos of AI speaking on your behalf. You can quickly make professional-quality AI human videos without a camera, studio, voice actor, or filming crew. Users just enter a script and click a few buttons on screen to create a video of a lifelike AI avatar speaking those words. You can create videos as intuitively as writing a document, without learning complicated video editing software. Choose a language and voice, and the avatar reads the script naturally, with no need to hire a professional voice actor or announcer or record it yourself.
Hee-ju) Several people work behind the scenes, each handling different parts of the service, so users can create videos easily with just a few clicks. These days, we're fixing pain points one by one, guided by the question, “Can someone new to STUDIO PERSO make a video easily without getting lost?”
Q. What process goes into creating the AI avatars used in the videos?
Hee-ju) For a Clone Avatar modeled on a real person, we film them in the studio in our office basement. Rather than filming every action, we capture poses such as front and side views, along with a range of movements, and use that data to create an avatar. We train the AI on speaking videos to produce more natural lip-sync. Once the resulting avatar assets are delivered to our team, we register the avatar information on the STUDIO PERSO server so customers can use it.
Sang-jin) When a company needs its own unique avatar, we create a branded avatar through a single filming and AI training process like this. Once it's made, you can use it on the STUDIO PERSO server to create videos with the desired movements and speech, without any additional filming. If you don't need a unique branded avatar, you can also make videos with our virtual character avatars.
Q. Where are these videos commonly used?
Hee-ju) They're mainly used for corporate communication videos, such as training materials, internal announcements, and product explanations. Companies turn the messages they want to share into videos presented by an AI avatar.
Sang-jin) The most common use is corporate training and lectures. It used to take several days to make an internal training video, from finding an instructor to filming and editing. Now, just enter a script and a video of an avatar instructor explaining it is ready. Upload a PowerPoint and the video is automatically organized into scenes. Even if the lecture content changes, you only need to edit the script instead of filming again.
The key to using STUDIO PERSO is that you can “make videos featuring people without filming.” It saves the time and cost of filming, makes it easy to revise content, and isn't affected by the talent's schedule. That's why the more often and in greater volume a company needs to make videos, the more it can feel the impact.
Q. Reducing the resources required for video production is a huge advantage. There are many AI video creation tools these days. What sets STUDIO PERSO apart?
Hee-ju) As I mentioned earlier, its biggest strength is the ability to create professional videos featuring people without a camera, studio, voice actor, or filming crew. Just enter a script, and the AI avatar effectively takes care of filming and editing. Even when the content changes, you only need to edit the text, making it especially suitable for content that needs to be created and revised often, such as corporate training, lectures, and informational videos. It offers more than 100 templates and over 6,000 voices to ease the burden of planning and designing videos. High-quality 4K videos are also easy for anyone to create, just like writing a document. Another strength is that ESTsoft's proprietary technology lets us provide an integrated solution, from avatar creation to video editing. ESTsoft films models directly at its AI human studio and trains them with deep learning to create avatars. Another point of distinction is that we can provide custom, unique branded avatars using our own avatars, without portrait rights or copyright issues. Our wide range of Korean avatars, well suited to the needs of Korean businesses,is another strength.
Sang-jin) As you said, there are a lot of tools for creating videos with AI these days. But I think what sets us apart is continuity, in two senses. The first is continuity over time. Most generative models create short clips lasting a few dozen seconds, but STUDIO PERSO generates scenes by script, so you can create a 10-minute presentation video. If you want to change just one sentence in a long video, you can edit only that part. The other is continuity between videos. Even if a company creates 200 videos of different types, it can maintain the same face, voice, consistent brand tone, and pronunciation. This kind of consistency is difficult to achieve simply by improving the generative model's performance, so it's something we consider important over the long term and continue to improve.
There are many AI tools for creating videos, but what makes STUDIO PERSO special is that it goes beyond simply making good videos. The STUDIO PERSO team considers both the convenience users experience firsthand and the quality of the finished product. Our goal is to provide a service that anyone can use to make videos easily, that reliably delivers the expected quality whenever needed, and that can flexibly accommodate revisions.
Powered by ESTsoft's AI technology, STUDIO PERSO is creating an environment where content can be produced more efficiently while reducing the time and cost of video production. By continuously improving the user experience, we also plan to strengthen our competitiveness in the global market as an AI service that transforms how companies produce videos. To learn more about STUDIO PERSO, visit the official website.
In the next installment, we'll take a closer look at the people behind STUDIO PERSO. From the challenges they face while developing the product to the future the team is shaping together, we'll hear the stories of the two developers behind it.
![[ITself] Delivering a Consistent Message with Easy-to-Make AI Avatar Videos: Perso Studio ① Attached Image 1](https://framerusercontent.com/images/w7dBWr8Q4i47Dn3tgmjMb7NLfHU.png)
It took me less than five minutes to try STUDIO PERSO for the first time, enter the interview script, and make an introduction video.
![[ITself] Delivering a Consistent Message with Easy-to-Make AI Avatar Videos: Perso Studio ① Main Image](https://framerusercontent.com/images/wKvL35vWLvGwbNwz8WqRCbenQI.jpg?width=1920&height=1080)
![[ESTsoft] ESTsoft provides Seven Eleven with its intelligent AI avatar “AI Mate” to test its usefulness in convenience store shopping!](https://framerusercontent.com/images/N9N0W8oRGGl0nZWld0DSqKJsdqs.webp?width=1440&height=900)
![[ESTsoft] Real-Time AI Avatar Service Perso Interactive Expands Across Public Services, Building Customized AI Services for Local Governments and Public Institutions](https://framerusercontent.com/images/Iie6Ol3RQYNZcXd8NsouLLX7TT8.webp?width=900&height=502)
