The AI told me it was impossible. But you don't have to give up there

That was the first thing I thought
when I listened back to my recorded voice 😅

I am currently working on
my third Udemy course.

For the narration of that course,
I am having my voice clone
read the script.

Once you register your voice,
all you have to do is type the text
and the system makes that voice speak it.

I'm using an audio service called Fish Audio,
and it's much more accurate
than I expected.

In other words, if the original audio isn't good,
even those flaws
are reproduced exactly as they are.

At first, I recorded it using the
Voice Memos app that comes with the Mac.

I thought I spoke normally
and recorded it normally, but...

The resulting voice
had a faint layer of noise on it.

The noise from the original recording
remained in the voice just like that 😨

So, I asked the AI.

I immediately tried asking the AI.

💭 Isn't there a way to create clean, noise-free audio for free?

The response I got back
was basically this.

In other words,it's impossible in your current environment
is the answer.

Hmm, I see.

Well, I figured that would happen.

I certainly did think
that I wanted a good microphone.

I also thought it was necessary
to deliver something better.

But
at this very moment, there is no microphone in front of me.

In that state,
I want to find a way to make it work.

I remembered the time I made my first course.

I was groaning in frustration for a while,
then I suddenly remembered.

When I made my first Udemy course,
I was recording the video
using Zoom's recording feature.

And I was removing the noise from that video's audio
using a service called Vrew.

Couldn't I just do the same thing
with just the audio?

I tried it, and it disappeared.

I only did three things.

1. Open a meeting by myself on Zoom and record it

2. Import the resulting audio into Vrew and remove the noise

3. Upload that audio and register my voice

The result was
surprisingly clean✨

The background noise was almost gone,
and the voice clone created from that
naturally became clean as well.

By the way, I tried
a total of four different recording methods.

While all of them resulted in
the upload assessment saying
"the speech is somewhat unclear,"

the only one that came back as "good"
was this Zoom to Vrew combination.

Why does it work?

Since Zoom is a service for calls,
it has built-in processing to suppress
room echo and microphone noise from the start.

So, by the time you record,
the echo and noise are
already reduced.

The AI didn't tell me this.

This is the point I most wanted
to write about today.

The AI never once
suggested this method.

But I don't mean to say
that the AI is wrong.

(Though it does make
huge mistakes sometimes, haha)

Telling you to use a good microphone
is a truly correct answer.

However, I realized that the ingenuity
people find by trying various things
within their own environments
exists outside of the AI.

Even Vrew itself
was something I learned about
from a teacher who teaches online business.

As for Zoom,
I really just happened
to stumble upon it myself.

Neither of these
came from asking an AI.

That's why gathering information

Thinking about it that way,
I think gathering information is important.

I try to pick up information on X every day
that I can apply to my own work.
I try to pick up information every day.

Having that allows me to
respond to AI's answers.

💭 No, I heard there's a service like this recently

I can talk back like that.🔥

When AI says it can't be done,
instead of just giving up,
I try to stick with it a little longer.

I believe that the information I've accumulated
is what supports that persistence.
I believe that.

The answer you get after persisting becomes your own.

I thought of one more thing.

The fact that you squeezed out an answer
that didn't come from AI means

that it becomes
your own unique experience.

It doesn't come up even if you search for it.

It doesn't come up even if you ask AI.

But there are definitely people
struggling with the same thing.

If so,
that experience becomes content
that you can pass on to someone else.

Today's article
is exactly that.

The story of being told it was impossible by AI
and finding a way through it by persisting
has become this article.

A method I found myself
after asking AI and getting no results.

I'm sure you have
something too.

Assuming your own experience becomes the content,
how do you deliver it?

Getting your customers
to know about it,
to get interested in it,
and to receive it.

This sequence
hasn't changed in 100 years.

The only thing that has changed
is the tools used to shape it.

So, with today's tools,
what should you create and in what order?

I have created
that roadmap.

If you have this first,
you can turn the experience you found through persistence
directly into the content of your product.

Feel free to take a look if you'd like 😊

🟠 Claude course, now available!

I have compiled the method
for building such a system
together with Claude into a course.

I am giving it away for free right now.
Feel free to take a look if you'd like!

You can receive it here 👇

👨‍💻 Click here if you're wondering who Taka is

👇 Recommended articles