I Gave the Same K-Drama Topic to 3 AI Video Podcast Generators — Here’s What Happened
Why are K-dramas so addictive?
Maybe it’s the slow-burn romance. Maybe it’s the emotional storytelling, characters you get strangely attached to, perfectly timed soundtracks—or those cliffhangers that convince you to watch just one more episode at 2 AM.
Whatever the reason, K-dramas give people plenty to talk about.
And that made me wonder: what would happen if I gave the same K-drama topic to three AI video podcast generators?
Instead of comparing feature lists and pricing pages, I wanted to actually use the tools.
So I tested HeyGen, JoggAI, and PodcastorAI with the same topic:
Why Are K-Dramas So Addictive?
I explored their scripts, AI hosts, voices, customization options, free-plan restrictions, and—where possible—the final generated videos.
I was also curious about how natural these AI-generated podcasts would actually feel, something I explored earlier when looking at how to use an AI podcast generator without sounding robotic.
I wasn't trying to find a winner. I wanted to see how differently three AI tools would turn the same K-drama idea into video content.
And the experiences ended up being much more different than I expected.
Table of Contents
One K-Drama Topic, Three AI Video Podcast Generators
To make the experiment as consistent as possible, I kept the main idea the same across all three tools.
I wanted the podcast to talk about some of the things that make K-dramas so binge-worthy: emotional storytelling, romance, relatable characters, cliffhangers, and memorable soundtracks.
I also used the free version or free trial of each platform.
While testing, I paid attention to a few things:
How easy was it to understand the creation process?
Could the tool help generate the script?
What could I change about the hosts, avatars, and voices?
How natural did the AI presenters look?
What could I actually create for free?
Were there watermarks, credit restrictions, or duration limits?
And most importantly: could I actually generate the final video?
That last question became more important than I expected.
1. I Started With HeyGen
The first tool I tested was HeyGen.
I started with the same K-drama idea and used its script-generation option rather than preparing everything beforehand.
One thing I noticed quickly was how structured the creation experience felt.
The process guided me through different stages, including the script, delivery, avatar, and video. As someone exploring the tool for the first time, this made it easier for me to understand what I was supposed to do next.
My K-Drama Script Had to Get Much Shorter
I initially wanted the script to cover several reasons K-dramas are so addictive—emotional storytelling, relatable characters, romance, cliffhangers, and soundtracks.
But in the free workflow I tested, I encountered a 210-character script limit. My generated script went beyond that limit, so I had to shorten it.
Eventually, I brought it down to a much smaller version that focused on the main ideas.
This changed the experiment slightly. Instead of seeing how HeyGen handled a longer K-drama conversation, I was seeing what it could do with a very short piece of content within the free workflow.
Avatar Customization Was One of the Most Interesting Parts
This was where HeyGen became more fun to experiment with.
Rather than simply accepting one default presenter, I could explore different avatar options and customize how the presenter would look.
I found options around characteristics such as appearance, age, gender, ethnicity, pose, orientation, and style. For a topic like K-dramas, where the presentation itself can affect the mood of the content, having more control over the presenter was useful.
The voice-delivery controls also gave me more to work with. I could explore the voice, adjust the speed, work on pronunciation, and review how the content would be delivered.
There was also built-in AI help, which I found useful when I wanted to understand something without leaving the platform.
What Happened When I Generated the Video?
I was able to generate the video using the free plan.
During my test, the free plan included 3 videos per month, with videos limited to up to 1 minute each.
The free output was 720p and included HeyGen branding/watermarking, while higher-quality 1080p output and watermark removal were associated with an upgrade.
The watermark was quite noticeable to me in the final result, particularly because of where it appeared in relation to the presenter.
But there was another side to the final output that stood out:
The avatar's expressions were quite good.
The presenter didn't just look like a completely static image speaking over audio. Compared with another tool I tested later, I noticed more facial expression in the HeyGen output.
So my HeyGen experience had an interesting trade-off: I liked the avatar presentation and guided creation process, while the free-output restrictions were very visible in the final video.
2. Then I Tried Turning the Same K-Drama Idea Into a JoggAI Video Podcast
Next came JoggAI.
This time, I specifically explored its Video Podcast workflow.
I again used the same idea:
Why Are K-Dramas So Addictive?
The setup allowed me to work with a podcast-style format using two AI presenters, which made the content feel closer to a conversation than a single-presenter video.
I could work on the script, explore the presenters, and prepare the podcast before rendering. And then I reached the part that made this test particularly interesting.
I Had 200 Credits. My Podcast Needed 300.
My free JoggAI account showed 200 credits.
When I reached the rendering stage, my podcast required more credits than I had available.
So I tried something logical:
I shortened the script.
I wanted to see whether making the podcast shorter would bring the credit requirement within my available balance.
But even after shortening it, the final rendering screen showed:
Credits needed: 300
My balance: 200
So, I could explore the Video Podcast workflow, work with the script and presenters, and prepare the content—but I couldn't complete the final render with the credits available in my account.
This was probably one of the most useful things I learned from experimenting myself.
If I had only looked at the pricing information and seen “200 credits,” that number wouldn't have told me how far those credits would take me.
Actually attempting to generate the K-drama podcast gave those credits some context. In my particular test, 200 available credits weren't enough to complete the Video Podcast I had created.
Because I couldn't render the final video, I also couldn't fairly judge the final video quality, facial expressions, or finished viewing experience.
And that's important to mention. I don't want to compare a final output I never actually saw with videos I successfully generated elsewhere.
3. Finally, I Gave the K-Drama Topic to PodcastorAI
My final test was PodcastorAI.
By this point, I had already experienced two very different workflows, so I was curious about what would happen with the same topic here.
PodcastorAI immediately felt focused on podcast creation.
The interface offered several formats, including Talk Show, Solo, Split Screen, and Two-Shot, along with other presentation options.
For this experiment, I wanted two speakers because it felt more natural for a topic like:
Why Are K-Dramas So Addictive?
Instead of one presenter simply explaining the answer, two AI hosts could discuss it with each other.
It Generated the Conversation for Me
I used the script-generation functionality to turn the K-drama idea into a conversation between two hosts.
The script was divided between Host 1 and Host 2, which made it easy to see how the discussion would move back and forth.
I could also explore different host pairs and presentation layouts.
One difference I noticed compared with HeyGen was host customization.
PodcastorAI gave me different existing hosts to choose from, but during my test, I couldn't find an option to specifically choose the ethnicity of the speakers.
That's not necessarily a problem for every use case, but it was something I noticed because HeyGen had given me more detailed control over avatar characteristics.
Then I Hit the 2-Minute Trial Limit
The free trial gave me 50 credits.
The trial information showed approximately 12 minutes of audio or 5 minutes of video creation, but individual trial videos were limited to a maximum of 2 minutes.
My generated K-drama conversation was longer than that.
Instead of simply stopping me, PodcastorAI displayed a message explaining that the current plan had a two-minute limit and that content beyond the limit would be trimmed.
I could either continue with the shortened output or upgrade.
I liked having that information before proceeding because I knew what would happen to the content.
And This Time, I Got the Final Two-Host Video
Unlike my JoggAI test, I was able to reach the final output.
Rendering wasn't immediate. During generation, the platform showed an estimated wait of more than five minutes and presented an upgrade option for faster rendering.
Eventually, though, my K-drama video podcast was ready.
And visually, I thought the video quality looked good.
The podcast had two presenters in a realistic studio-style environment, and the camera perspective changed depending on the speaker.
It looked much more like the type of video podcast format I had originally wanted to test.
But I noticed something else when I watched the hosts speak.
Their facial expressions felt relatively limited to me.
This became especially noticeable because I had already watched my HeyGen output. In my test
HeyGen avatar appeared more expressive, while PodcastorAI gave me a clean-looking two-host podcast format but with less expressive presenters.
PodcastorAI also displayed its branding on the output, although I personally found the branding less distracting than the watermarking I encountered in my HeyGen test.
That doesn't make one approach automatically better than the other—it was simply one of the visual differences I noticed when watching my generated videos.
HeyGen vs JoggAI vs PodcastorAI: What Actually Happened?
Once I finished testing, this was the simplest way to summarize my experience:
This is also why I don't think looking for a simple “best AI video podcast generator” tells the whole story.
My experience changed depending on what part of the process I was looking at.
What Testing a K-Drama Topic Taught Me About AI Video Podcast Generators
I originally thought the biggest difference between these tools would be the quality of the AI-generated script.
It wasn't.
Once I actually tried creating something, four other things became just as important.
1. “Free” Means Something Different on Every Platform
This was probably my biggest practical takeaway. HeyGen gave me a limited number and duration of free videos.
JoggAI gave me credits, but the Video Podcast I created required more credits than my available
balance.
PodcastorAI gave me trial credits but limited individual trial videos to two minutes.
All three offered some form of free access, but what I could actually produce with that access was very different. That's why I'd now look beyond the word Free on a pricing page and ask:
Can the free allowance actually complete the type of video I want to create?
2. Facial Expressions Matter More Than I Expected
Before this test, I was mostly thinking about script quality, voices, and video quality. But creating a natural-feeling AI podcast involves more than getting the voice right—delivery and presentation can affect how human the final content feels, too.
Then I watched the presenters.
For conversational content—especially something emotional and fun like K-dramas—the presenter can change how the whole video feels.
During my testing, HeyGen's avatar appeared more expressive to me.
PodcastorAI's final video looked clean and had a convincing podcast setup, but I noticed less facial expression from its presenters.
That difference made me realize that realism isn't only about how realistic an avatar looks in a still image.
Movement, expression, and delivery matter too.
3. Watermarks Can Affect the Viewing Experience
Free AI tools understandably need to differentiate free and paid output, and watermarks are one way of doing that. But where the watermark appears matters.
I found HeyGen's watermark particularly noticeable in my generated video.
PodcastorAI also included its branding, but I found it less distracting while watching the conversation.
If you're testing these tools before paying, this is something worth checking in the actual output—not just on the pricing page.
4. Credits Don't Mean Much Until You Try to Generate Something
JoggAI demonstrated this particularly clearly for me.
Seeing 200 credits initially sounds useful.
But I didn't know whether 200 credits meant one complete podcast, several short videos, or only part of the workflow.
Once my actual K-drama Video Podcast required 300 credits, the number suddenly had context.
So when comparing credit-based AI tools, I'd pay attention not only to how many credits you receive, but also to how many credits your actual type of project consumes.
Why I'm Glad I Used a K-Drama Topic for This Test
I could have tested all three platforms with something generic like:
“Five productivity tips for entrepreneurs.”
But I'm glad I didn't.
Using “Why Are K-Dramas So Addictive?” made it easier to judge what the AI was producing because I already knew what kind of conversation I expected.
Choosing a topic people already care about can also make podcast content easier to turn into a conversation rather than simply presenting information.
There should be emotion. There should be discussion around characters, romance, storytelling, cliffhangers and music.
And because K-dramas themselves rely so much on emotion and human expression, the topic unexpectedly made avatar expressions more noticeable too. A presenter discussing emotional K-drama moments while showing very little expression feels different from a presenter delivering the same content with more visible emotion.
So the K-drama angle wasn't just a fun topic for the experiment. It actually helped me notice differences between the AI presenters that I might have overlooked with a more technical topic.
So, Which Experience Was Different for What Reason?
After testing the same K-drama topic across all three platforms, I wouldn't reduce the experiment to one tool being “good” and another being “bad.”
They gave me different experiences.
The biggest thing I took away from this comparison is that these differences matter because creators aren't necessarily looking for the same thing.
One person might care most about avatar expression and customization. Someone else may want a traditional two-host podcast format. Another creator may care primarily about how much they can actually produce with the free allowance before needing to upgrade.
That's why I don't think this experiment comes down to simply calling one experience “good” and another “bad.” The more useful question is which differences matter most for the kind of video podcast you're trying to create?
Final Thoughts: Same K-Drama Topic, Three Very Different Experiences
I started this test with one simple question:
What happens when I give three AI video podcast generators the same K-drama topic?
I expected similar results, but the experience was quite different across HeyGen, JoggAI, and PodcastorAI.
The biggest lesson for me was that feature lists and pricing pages only tell part of the story. It's only when you actually create something that you notice things like credit limitations, avatar expressions, watermarks, and how the final video feels to watch.
So if you're comparing AI video podcast generators, try giving them the same topic you'd genuinely want to create content about and go through the actual workflow.
If you want to try creating one yourself, you can also experiment with ALwrity's Podcast Maker and see how your own topic turns into a podcast.
For me, that topic was K-dramas—and it turned out to be a pretty useful way to see what these tools actually feel like to use.
🎬 Before You Go… One K-Drama Question
We used “Why Are K-Dramas So Addictive?” to test three AI video podcast generators. Now I want to know what keeps you watching.
What usually makes you say “just one more episode”? 👀
Vote and see what other K-drama fans picked. 🍿
Comments