A 10-second version of "Au Clair De La Lune," now archived online, is said to predate Thomas Edison's Manhattan railroad sounds.

K.C. Jones, Contributor

March 28, 2008

2 Min Read

A group of researchers has played what is thought to be the oldest recording of a human voice.

The recording played Thursday predates Thomas Edison's invention of the phonograph (previously thought to have recorded the first sound) by 17 years. It captured about 10 seconds of the French folksong "Au Clair De La Lune" on April 9, 1860.

Édouard-Léon Scott de Martinville recorded the voice by using a "phonautograph" to scratch sound waves onto a sheet of paper covered in black smoke from an oil lamp. He never intended to play the sounds. Instead, he archived the recording and patented a method for understanding sound. Researchers recently unearthed the recording at the Academy of Sciences (French) in Paris.

Audio historians, recording engineers, and scientists working in conjunction with the informal collaborative group First Sounds created high-resolution, high-grade scans of Scott's phonautogram, converted the images into digital form, and played the sounds on a computer with a virtual stylus. Then they evened out speed fluctuations and tweaked the tracks to pull the voice forward.

First Sounds historians Patrick Feaster and David Giovannoni began looking for surviving phonautograph recordings, or phonautograms, just a few months ago. In October, they studied 19 recordings that Edison made in 1878. Edison and his associates created the recordings to study the noise of the Metropolitan Elevated Railroad in Manhattan, and the Edison National Historic Site in New Jersey stored them.

In December, the researchers went to French patent offices and identified two phonautograms that Léon Scott submitted with patent applications from 1857 and 1859. In February, they confirmed that nearly a dozen of Scott's phonautograms survived at the Academy of Sciences. They included Scott's earliest experiments from 1853 or 1854, as well as more advanced recordings from 1860.

Earl Cornell and Carl Haber, scientists at Lawrence Berkeley National Laboratory, converted the scanned etchings into digital sound, using technology that preserves a wide variety of early recordings on grooved mechanical carriers, such as phonograph discs and cylinders. Studio engineer Richard Martin adjusted the speed of the song and teased out the voice from scratchy background noise.

The researchers premiered the recording at the annual conference of the Association for Recorded Sound Collections at Stanford University on Thursday.

About the Author(s)

Never Miss a Beat: Get a snapshot of the issues affecting the IT industry straight to your inbox.

You May Also Like


More Insights