8/24/2024

How AI is Bringing JFK’s Voice Back to Life

Artificial Intelligence (AI) is making headlines for a multitude of reasons these days, but one of the most intriguing developments is that it is reviving the voices of icons from the past. One prominent example is the voice of John F. Kennedy (JFK), the 35th president of the United States. With AI enhancements, we can hear JFK's voice once again—particularly a speech he was due to deliver on the day he was tragically assassinated. In this post, we will explore how this technological marvel works, the implications of such endeavors, and how you can also engage with AI technologies at Arsturn.

The Dream of Recreating JFK’s Voice

On November 22, 1963, President JFK was set to deliver a speech at the Dallas Trade Mart, a moment that was tragically cut short by his assassination. For decades, this 'lost' speech remained unheard, sparking the imagination of many who pondered what it would have contained. Fast-forward to 2018, when a collaboration between the Times newspaper and the creative agency Rothco leveraged the advancements in AI technology to recreate that speech using JFK's own voice.

Technology Behind the Voice Resurgence

The journey to recreate JFK’s voice involved complex AI technologies, particularly through voice synthesis. The project utilized the expertise of CereProc, a tech company specializing in text-to-voice synthesis technology. This method involved analyzing and stitching together numerous audio samples of JFK's recordings. Here’s how they did it:

Data Collection: The team reviewed over 831 of JFK's speeches and interviews, isolating elements of his voice including tonality, pitch range, and characteristic phrases.
Digital Synthesis: They analyzed 116,777 sound units and crafted a computer program capable of synthesizing individual phonemes that resemble his actual speech patterns. This created a digital library from which the AI could draw.
Deep Neural Networks: The technology employed complex algorithms to harmonize the speech in a natural-sounding way, adjusting for emphasis and cadence that made JFK’s speaking style unique.
Post-Production: Vocals were perfected through audio engineering processes so that they sounded as lifelike and engaging as possible.

The Unveiling of JFK’s Speech

Upon completing the intricate process, the Dallas Trade Mart speech was finally made available to the public. Barry Pidcock, a co-founder and chief voice engineer at CereProc, asserted that the team had never attempted such a project on this scale before. The result was both profound and reflective of JFK’s notable eloquence. The speech highlighted issues of freedom, power, and wisdom, emphasizing that *