Case Study: A Nike Commercial Brought to Life with AI

In this article, we delve into an exceptional case study: creating a voiceover for a Nike commercial using AI technology. This wasn’t just a choice—it was a necessity. And, as we discovered, AI worked like a charm. We’ll explore why AI was the ideal solution, what exactly we did, and how we did it.

The Pivotal Role of Audio in Video Production

It’s an undeniable fact: the right audio and speech contribute to over half of a video production’s success. In the dynamic world of content creation, audio quality and voiceover work are not just complementary elements but rather the backbone of storytelling. The evolution of AI technologies has further expanded these possibilities, bringing us closer to limitless potential in both content creation and audio production.

The Story Behind the Commercial

Our spec commercial for Nike centered around the compelling story of Qleem, an immigrant and refugee who fled Russia amidst the turmoil of the Russian invasion of Ukraine. His journey is one of resilience and survival, marked by protests, a lost home, and ultimately, a daring escape to avoid unjust imprisonment.

Qleem’s story doesn’t end with his flight; it begins anew in the United States. It’s here, on a basketball court, that he battles loneliness, depression, and the shadows of his past. Basketball becomes his refuge, and through it, he forges a friendship that sets him on the path to finding a true sense of home.

Qleem’s Voice: A Character Defining Element

Qleem’s personality is unique, and so is his voice. It’s distinct, imbued with a charming accent that perfectly encapsulates his journey and character. This voice, we believed, was integral to our commercial’s narrative—it had to be preserved in its most authentic form.

The Challenge of Authenticity and Pace

But here’s the catch: while Qleem diligently worked on his English language skills, his proficiency didn’t align with the fast-paced nature of our commercial. No matter how hard he tried, his delivery couldn’t match the dynamic tempo we envisioned.

Enter AI: The ElevenLabs Solution to Our Voiceover Challenge

This is where the magic of AI speech generation came into play. We turned to ElevenLabs’ cutting-edge voice generation tool, not just as a workaround, but as a strategic choice to maintain the authenticity of Qleem’s voice while adhering to the commercial’s dynamic rhythm. The tool offered us a unique opportunity to create a voiceover that was true to Qleem’s distinct accent and personality, yet polished enough to fit seamlessly into the high-energy flow of the commercial.

Choosing the Right AI Tool: Why ElevenLabs Stood Out

In our quest for the perfect AI-based speech generation tool, ElevenLabs emerged as the clear frontrunner. Although their free version comes with limited features, it provides ample opportunity to test the product’s compatibility with specific video production and voice generation needs. For our project, the Starter plan, their most affordable option, proved more than sufficient. However, achieving our goal wasn’t without its challenges.

Mastering ElevenLabs Voice Generation Feature

The first step with ElevenLabs involves recording a one-minute speech piece. We asked Qleem to provide this sample. Utilizing this recording, we experimented with the Text-to-Speech feature, adjusting various settings like Stability, Clarity, and Voice Exaggeration. Despite our efforts, the initial attempts fell short of our expectations. The main issue was the speech speed—it simply couldn’t match the brisk pace of our video. Just when we thought we had hit a dead end, a breakthrough approach turned the tide.

A Creative Solution: Blending Human Effort with AI Precision

I personally took on the task of reading and recording the voiceover text at the required pace, infusing various tonalities and stress levels to ensure flexibility in post-production. Remember, I’m no professional voice artist, so this step required some experimentation.

Then, employing the Speech-to-Speech option and selecting Qleem’s voice from the library, we transformed my recording. The result was astonishing. The AI-generated voice was clear, natural, and far from robotic. It was as if we were hearing Qleem speak with a few more years of English practice under his belt, yet retaining his unique tone and accent.

The Role of AI in Modern Video Production

In reflecting on this experience, it’s clear that AI tools, while advanced, are not poised to replace full-service video production teams. The artistry in cinematography, photography, and storytelling remains irreplaceable and gains a premium status, akin to the reverence for books, vinyl records, and other classic mediums.

However, our experience with ElevenLabs signifies a milestone. It marks the first time an AI tool has not only met but exceeded our expectations in a commercial project. Going forward, we anticipate incorporating ElevenLabs’ speech generation technology as a staple in our arsenal for commercial, corporate, and promotional video productions.