Remember when we thought it was impossible to have Audrey Hepburn or John F. Kennedy speak in this era? The development of synthetic voice technology proved us wrong. Now, we can have our favorite celebrities and historical figures "speak" again, thanks to AI’s innovative synthesis. It’s a fascinating progression in technology that benefits many industries – from entertainment to education, and even healthcare.
Celebrity voices revived with AI
It’s been a while since we could hear the iconic voices of James Dean, Marilyn Monroe, and Elvis Presley. They have now passed away, but that doesn’t mean they can’t come back to life with the help of AI. Through machine-learning algorithms, companies like Veritone and VocaliD can generate synthetic voices of well-known personalities for commercial use or philanthropic causes.
The rise of synthetic voice technology
The advancement of speech recognition and text-to-speech technology paved the way for synthetic voices. Speech recognition software identifies the patterns and traits of a voice, while text-to-speech technology can convert electronic texts into speech. Synthetic voice technology combines both to make an almost indistinguishable synthetic voice from human speech.
This technology isn’t limited to reviving celebrity voices; it’s also helpful for people with severe speech impediments, motor neuron disease, and amyotrophic lateral sclerosis (ALS). People diagnosed with ALS, for instance, struggle with communication because of paralysis. With synthetic voice technology, they can have a personalized voice output for efficient interaction in their daily lives.
A glimpse into innovative synthesis
The process of recreating a celebrity voice or creating an entirely new one follows some steps. First, voice actors or speakers record hours of speech data. The data is then used to train machine-learning algorithms that can replicate the sound and pitch of the voice. Lastly, specialists fine-tune the algorithm to add pinpoint intonations and inflections. The result is an almost perfect synthetic voice, with the same emotional tone and timbre as a human voice.
Besides recreating a celebrity voice, AI technology is capable of mixing two or more voices to create a new one. The Synthesia project, for instance, uses a video of a person speaking and synthesizes the voice by mapping the characteristics of the speaker’s mouth, lip motion, and facial expressions. The technology can produce a vast range of synthetic voices tailor-made to different contexts, like the ages, genders, and accents of speaking characters in a video game or animation.
Hear your favorite stars speak once more!
With the abundance of synthetic voice technology, it’s now possible to hear a variety of popular figures speak. For instance, Morgan Freeman has partnered with software company WAVi for a project that generates real-time dialogue in his voice. Fans of Star Trek can use the Waze navigation app to hear the voice of Captain Kirk, also synthesized through AI. The AI program, VoCo, even creates a synthetic voice for Steve Jobs’ biography audiobook narration.
Moreover, AI technology has enabled marketers to create interactive ads using synthetic voice technology. Adobe’s Project VoCo is one of the tools used for generating synthetic voices for commercial purposes. And in education, researchers are utilizing the technology to preserve the voices of Holocaust survivors and other historical figures for generations to come.
AI’s innovative synthesis of celebrity voices gives us new ways to appreciate the sounds of our past and present idols. The technology is changing the way we can interact, learn, and entertain ourselves. As AI continues to develop, so might our opportunity to explore all the possibilities of synthetic voice technology.