Vocal Codes
Every time you make a digital cell-phone call, you take advantage of speech coding-the process of converting human speech into a simpler signal that can be transmitted more quickly. This technology has its roots in a quirky organlike electrical speech synthesizer invented nearly 70 years ago by an AT&T researcher.
In the mid-1930s, Bell Labs electrical engineer Homer Dudley was attempting to create a speech synthesizer, hoping it could reduce the amount of bandwidth necessary to transmit speech over transatlantic telegraph cables. Dudley knew that sounds produced by the vocal cords are transformed into speech by small changes in the shape of the space inside the mouth, which amplify certain frequencies and damp others. Dudley realized he could measure these frequency variations by passing spoken words through a bank of 10 filters that gauged energy output over different portions of the audio spectrum. By applying the same variations to electronically generated tones, he was able to produce robotlike but intelligible speech. His device, the Vocoder, was a breakthrough, enabling speech transmission that used far less bandwidth than sending an entire speech signal directly, which is how telephones worked at the time.
Still, Dudley’s device didn’t get much attention until he showed a revised version, dubbed the Voder, at the New York and San Francisco World’s Fairs in 1939. Instead of simply speaking into the device, a trained technician-usually a young female telephone operator-would shape tones into synthetic speech by simultaneously manipulating a series of keys and a foot pedal. The Voder made quite a splash at the fairs, but it didn’t turn its predecessor, the Vocoder, into a widely used bandwidth-reducing device. Undeterred, Dudley found another use for the Vocoder a few years later-as a voice encryption device for defense communications during World War II.
Although Dudley’s research didn’t make him a household name, his work forms the foundation of the speech coding used in modern digital telephony. Edward Lee, a professor of electrical engineering and computer science at the University of California, Berkeley, learned about the Voder and Vocoder when one of Dudley’s colleagues from Bell Labs played him old recordings from the devices. “I didn’t realize that the concepts were as old as they were,” says Lee. “It really was a very significant development, and it deserves a lot more attention than it gets.”
Keep Reading
Most Popular
This new data poisoning tool lets artists fight back against generative AI
The tool, called Nightshade, messes up training data in ways that could cause serious damage to image-generating AI models.
The Biggest Questions: What is death?
New neuroscience is challenging our understanding of the dying process—bringing opportunities for the living.
Rogue superintelligence and merging with machines: Inside the mind of OpenAI’s chief scientist
An exclusive conversation with Ilya Sutskever on his fears for the future of AI and why they’ve made him change the focus of his life’s work.
How to fix the internet
If we want online discourse to improve, we need to move beyond the big platforms.
Stay connected
Get the latest updates from
MIT Technology Review
Discover special offers, top stories, upcoming events, and more.