I Dubbed One of My Videos Into Spanish. Total Cost: $0.

I dubbed one of my videos into Spanish. Total cost: $0.

Real pipeline, tested on a real 60-second video. Here's exactly what it took, what worked, what didn't, and the honest limits you need to know before you promise this to anyone.

Where the idea came from

A creator named Koerner runs an $8,000-a-month service dubbing videos into Spanish for local businesses and creators. He published his whole playbook: the five types of customers he targets, his pricing, even the exact DM he sends to get clients. He dubbed his own channel into 13 languages, and his reach and revenue grew with every one.

It's a great story. But I don't teach things I haven't tested myself. So I took one of my own Agent Files videos, a 60-second reel, and ran his playbook on it.

What I actually did

The pipeline has five steps. None of them require special software beyond a computer, free tools, and patience.

Step one: get the words and their timing. I pulled every spoken sentence from the video along with the exact second each sentence starts. Accuracy matters here, down to half a second. The best source is the video's own caption files if you have them. If not, transcribe it fresh.

Step two: translate for the ear, not the page. This is the step most people get wrong. Written Spanish and spoken Spanish feel different, just like in English. Numbers get spelled out. Sentences stay short. You're writing something a voice will say out loud, not something someone will read on paper.

Step three: generate the voice. One audio file per translated sentence, same voice all the way through. I used a free text-to-speech step for this. The key discipline: don't record one long file. One file per sentence. You'll see why in the next step.

Step four: re-time and mix. This is where the magic happens. A script takes each Spanish sentence and lines it up with the original sentence's start time. Spanish sentences are often longer or shorter than the English originals, so the script gently speeds up or slows down each one to fit. Anything stretched too far gets flagged for a human to check. Then it mixes the new voice under the original music.

Step five: verify, don't trust. The script checks its own work by analyzing the silences in the final audio. If a sentence drifts from its slot, the check fails and nothing ships. My first run took about 35 minutes end to end. Repeat runs take 15 to 20.

The honest limits

Here's where I pump the brakes, because this matters if you're thinking about selling this.

What I proved: voice dubbing works. The Spanish voice replaces the English voice, the timing lines up, it sounds natural. For a faceless video, narration over images, nobody can tell it wasn't always Spanish.

What I did not prove: lip-sync. That's the version where the speaker's mouth moves with the new language. Koerner's actual $8,000-a-month product includes talking-head videos, and that mouth-matching step is a different pipeline, usually a paid one. On free tools, it's unproven. I won't sell you what I haven't tested myself.

Two more gaps I found. The burned-in captions stay in English unless you re-render the whole video, so a Spanish viewer still reads English text on screen. And no matter how good the timing check is, a native Spanish speaker should listen to the final result before it goes to a real audience. Machines verify timing. Humans verify that it sounds like a person, not a robot reading a translation.

How to pick your first video

Not every video is a good dubbing candidate. Here's how I'd choose, based on what the test taught me.

Pick a video where the voice carries the meaning and the pictures just support it. Narration over footage, a story told over images, a lesson with visuals underneath. That's the sweet spot, because the voice is the only thing changing.

Skip talking-head videos for now. If someone's face is on screen the whole time, viewers will notice the lips don't match, and free tools can't fix that yet. That's the line between what I proved and what I didn't.

Skip videos where the words on screen do the heavy lifting. If your video has text overlays with the key points, jokes in the captions, or screenshots people need to read, the dub only translates half the experience. You'd need to re-render everything, which turns a 20-minute job into a real project.

And start with your best performer, not a random one. Dubbing doubles the audience of a video. Doubling a video nobody watched still gives you nobody. Take the one that already worked in English and give it a second life in Spanish. That's where the return is.

Why this is worth your time anyway

If you make videos for customers who speak Spanish, a free Spanish version of your best video is a real local wedge. Think about it: a roofer in Texas, a dentist in Florida, a restaurant anywhere. Their competitors are all English-only. One Spanish version of their best video, and suddenly they're the business that bothered.

The economics are silly. Twenty minutes and zero dollars to double the audience of a video you already made. Even with the limits, even with a native speaker check, that's the cheapest audience expansion I've ever tested.

I approved teaching this in my Skool community because it's real: tested on real content, verified objectively, limits stated plainly. That's the bar everything I teach has to clear.

The full playbook, including the pipeline script and the verification step, is in my free pack: https://infanodes.com/ai-employee.html