Why Are Some VOICEROID Travel Videos So Much More Entertaining Than Others?

Lately, I have been watching a lot of travel videos on YouTube that use VOICEROID characters.

They feature two synthetic-voice characters bantering as they introduce the scenery and food at their destination.

There is something curious about this genre.

Even when the footage, characters, and synthetic voices are similar, the videos can differ enormously in how entertaining they are.

I can watch some channels over and over and still enjoy them. With others, I get bored within five minutes.

Where does that difference come from?

The Difference Between a Travel Guide and Entertainment

In a dull video, the characters do little more than explain the destination.

“This is a famous tourist attraction.”

“This dish is a local specialty.”

“It was delicious.”

That is hardly different from listening to someone read a guidebook aloud.

In an entertaining video, by contrast, the banter between the characters takes center stage.

If one of them takes a wrong turn, the other makes fun of her. If they order an expensive meal, they worry about their wallets. If they came for a spectacular view only to be met by rain, they complain about it.

As travel information, it is almost useless.

But it is entertaining.

What viewers are enjoying is a two-person comedy routine built around a trip.

In Reality, One Person Is Writing Everything

What makes it even more interesting is that although two characters appear to be having a natural conversation on screen, a single creator is actually writing the entire script.

Take this familiar kind of joke, for example.

“Great! Luckily, there was exactly one seat left!”

“Wait. Aren’t there two of us?”

The screen suddenly cuts away.

“Please stand by.”

A few seconds later, the trip resumes as if nothing happened.

Of course, only one person—the creator—is actually traveling. One seat is all that is needed.

This kind of meta joke is genuinely funny.

And one person has come up with the setup, the comeback, and even the timing of the cutaway.

The creator plays the fool, delivers the retort, and edits the exchange—all alone. It is an impressively elaborate one-person performance.

The length of each line, the pauses, and the timing of each comeback: the better the video, the more carefully its script and direction have been crafted.

This is less like making a travel video and more like writing a comedy sketch.

The Same Tools Do Not Produce the Same Work

What is interesting about VOICEROID is that anyone can use the same characters and the same voices.

In other words, the people who make entertaining videos and the people who make dull ones are using exactly the same tools.

Even so, the finished results can be worlds apart.

In the end, it comes down to the sensibility of the person writing the script.

Even an ordinary destination can become entertaining when a funny person writes the script. Conversely, a rare and unusual destination will still produce a dull video if the script is boring.

In that respect, it resembles rakugo, the traditional Japanese art of comic storytelling.

The same story can vary enormously depending on the performer. A skilled rakugo storyteller can make it enjoyable again and again, even when you already know the plot.

VOICEROID travel videos probably work the same way.

The AI Era Will Produce an Even More Brutal Divide

This also applies to content created with AI.

Today, AI makes it easy to produce text, images, and audio. The barriers to video production will probably fall even further.

But making it possible for anyone to create something is not the same as making it possible for anyone to create something entertaining.

If anything, as differences between the tools disappear, differences in the creators’ sensibilities become more conspicuous.

In the past, people could make excuses: “My equipment isn’t good enough,” or “I don’t have the budget.” But when two people use the same software, the same characters, and the same voices, yet one result is entertaining and the other is dull, those excuses no longer work.

There is nowhere left to hide.

An age in which anyone can become a creator is also an age in which it becomes easy to see who lacks talent.

No matter how far AI advances, what will ultimately be tested is how interesting the human behind it is.

Most ironic of all, two characters can speak in exactly the same voice, yet only one of them sounds like a brilliant conversationalist.

Apparently, even when a machine does the talking, we cannot blame the machine for having nothing interesting to say.