Breaking the Language Barrier: AI and International Distribution

Most conversations I've heard recently about the application of AI in filmmaking have been focused on the incredible special effects that are now possible at a much lower price point. I've heard less about its impact on sound, but this is where, I believe, the “true magic of AI” exists, especially for independent filmmakers seeking distribution.

I just completed production of a documentary about the Portuguese island of Madeira, which was shot largely in Portuguese. I don't speak Portuguese. A few years ago, that fact alone would have meant handing the footage off to someone else for the edit. But by combining some new features in the editing software I use with AI's translation capabilities, I was able to edit the entire film myself, in a language I don't speak, following the meaning and rhythm of every interview with the same precision as if I were fluent in Portuguese.

Why This Mattered for Distribution

But that is only half the story. The worldwide film market is ranked by potential income, with the English market sitting at number one, while the European Portuguese TV market sits around #40 globally, with a maximum reach of less than 11 million viewers.

From the first stages of development this documentary was always intended for the English market, and interviews were conducted in English whenever possible, but often could only be recorded in Portuguese. Until recently, there were just a few ways to integrate those Portuguese interviews into an English-language film. Subtitles ask an audience to split their attention between reading and watching, but film doesn't work that way. Every second spent reading a line is a break from immersion in the world of the film as the story unfolds before you. Traditional dubbing solves that problem but creates a different one: when the voice track doesn't match the movement of the lips, something in the viewer's mind registers the mismatch. That small, constant friction breaks the suspension of disbelief that filmmakers work so hard to create. The interview stops feeling real, and starts feeling like a performance layered on top of a performance.

AI-assisted lip-sync closes that gap. It maps the subject's mouth movements to a new language track, so the visual performance stays intact while paired with realistic dialogue in a new language. The result is a film that plays naturally in a language the subject never actually spoke, without asking the audience to read subtitles or accept hearing words that don't quite match the speaker's delivery.

Our editorial approach also emphasized cinematic storytelling rather than long talking-head interviews. We established each subject on screen briefly, then let B-roll support the storyline while their voice continued underneath. The result was a more dynamic viewing experience, and it also minimized the need for lip sync. 

The use of AI closed the language gap twice over: first so I could understand and edit what was being said, then so audiences could hear it seamlessly in English.

As a result, the film can be efficiently adapted into German, French, Japanese, and other languages with minimal additional post-production.

This Is a Capability That Didn't Exist Until Just a Few Years Ago

The technology behind this traces back to 2020, when researchers first showed that a neural network could sync mouth movements to any audio track for any speaker, without training on that specific person. That was a research breakthrough, not a usable tool. The production-grade version, reliable enough to build a film on, only reached the market in 2023, when that same research team launched it as a commercial platform. What I'm using today is a further generation beyond that: higher resolution, better handling of multiple speakers, and translation built directly into the same workflow. So when I say this wasn't possible just a few years ago, I mean it. The capability I relied on for this documentary is only about three years old, and it's improving fast enough that what's possible now will look basic in another three years.

What This Means for Independent Filmmakers

For anyone considering a documentary, especially one that involves interviews conducted in multiple languages or distribution to international audiences, this changes the calculation considerably.

This is important when telling a story rooted in a specific place, culture, or community, where the people on screen may not speak the language of the audience you most want to reach. It means you're no longer forced to choose between authenticity on the shoot and reach in distribution. You can film the story the way it actually happened, in the native language, and still deliver it to an audience anywhere in the world as though it were made for them from the start.

A Recent Example of This in Practice

In 2025, the Swedish science fiction film Watch the Skies became the first international feature dubbed entirely with AI for a U.S. theatrical release, with the technology company Flawless AI handling both voice synthesis and lip sync using the original cast's own voices to produce an English-language version. A film that would have once been confined to festival screenings or a subtitled niche release reached mainstream American theaters instead, without the years-long dubbing budget that kind of theatrical crossover traditionally requires.

This isn't limited to major studio releases either. Streaming platforms are moving in the same direction. Prime Video recently launched a pilot program using AI-assisted dubbing on a group of licensed movies and TV series that had never been dubbed into another language before. The pilot covered twelve titles, adding English and Spanish audio tracks to content that previously only existed in its original language. The process uses a hybrid model: AI generates the first pass of the translated audio, and human localization professionals then review that output for quality and cultural accuracy before it's released. Some titles were previously considered uneconomical to dub. The high expense of traditional studio dubbing wasn't justified by the limited income from the small audience they'd reach in a new territory. Now that the cost structure has changed, those same titles are viable.

That's the real shift for independent and documentary filmmakers too. Traditional dubbing took a long time and cost thousands of dollars per language. AI-driven dubbing can cut production costs by 60 to 90 percent while reducing turnaround time by roughly 80 percent. A documentary that once had one realistic shot at a single-language broadcast now has a plausible, affordable path into markets that would have been out of reach a few years ago.

What This Means for Funding and Distribution

It also means something practical for how a production gets funded and sold. A distributor evaluating a documentary today isn't just asking whether the story is strong. They're asking how many territories it can realistically reach, and at what cost. While conversations about AI in filmmaking stay centered on what audiences see, I think it’s what they hear that could lead to financial success. A film that appealed to a niche audience in just one country may have failed in the past. But when that same film can reach small audiences in dozens of languages at once, those viewers could add up to success. The real benefit of AI isn't only the money it saves low budget filmmakers, but the money they earn by reaching a worldwide audience for the first time. That’s why I believe audio is the true "magic of AI." Where Do We Go From Here?

Artificial intelligence is changing documentary production faster than any other technology I've seen in my career. New tools appear almost weekly, and the opportunities and consequences of AI integration continue to evolve.

Throughout this blog series, I'll be sharing practical insights from projects where I'm integrating AI into real-world documentary production, from multilingual distribution and AI-assisted restoration to visual effects, production workflows, and the technologies that are affecting my work.

My goal isn't to chase every new technology. It's to identify the tools that improve storytelling while preserving the authenticity that makes documentaries so powerful.

If you're interested in how AI can expand what's possible without losing the human element, I hope you'll continue following the articles here at JDFreedman.net. I'll be sharing what works, what doesn't, and the lessons I learn as these tools continue to evolve.