Particle Physics Resurrects Alexander Graham Bell’s Voice

Source: IEEE Spectrum | April 30, 2018 | Allison Marsh

It takes some doing to extract sound from an 1885 wax disc

In 1880, Alexander Graham Bell decided to go head-to-head with Thomas Edison. His goal: improving—and commercializing—the phonograph. Bell established the Volta Laboratory in Washington, D.C., using prize money he’d received from the French government for his invention of the telephone. He hired his cousin, chemist Chichester Bell, and instrument maker Charles Sumner Tainter. Collectively known as the Volta Laboratory Associates, the three men spent the next five years researching the transmission and recording of sound.

Thomas Edison had invented the phonograph in 1877, successfully demonstrating how to record and play back sound. But there were problems. Edison’s machine recorded onto a thin piece of tinfoil wrapped around a cylinder. The foil tore easily, the grooves wore smooth after only a few plays, and the sound quality was poor. In short, nothing about the phonograph was practical. Bell saw an opportunity.

By 1885, the Volta Associates had applied for several patents for engraving sound on wax discs. They trademarked the name Graphophone to distinguish their product from Edison’s, and in early 1886 they incorporated the Volta Graphophone Co. to handle the commercial enterprises. One of their most successful products was the Dictaphone.

Bell knew there was a distinct possibility of a patent fight with Edison, so on three occasions he had papers and experimental products sealed in tin boxes and deposited at the Smithsonian Institution for safekeeping. They were to be opened only with the approval of at least two of the three Volta Associates. Although Bell never used the material as evidence in a legal battle, the Smithsonian collection grew to include hundreds of sound recordings and notebooks from the Volta lab.

These days, the Smithsonian, like many other museums, has to carefully balance the preservation of its collections with its ongoing research. Generally speaking, it’s not a good idea to destroy unique museum objects while on the quest for new knowledge, or to halt all research so as to preserve more objects.

Most of the Volta recordings at the Smithsonian are one of a kind. They cannot be played, either because the original playback apparatus no longer exists or because to do so would permanently damage the record. For years, curators thought they had lost this sonic cultural heritage forever, and the objects sat mutely in storage.

Then in 2003, science caught up with the recordings. Vitaliy Fadeyev and Carl Haber of the Lawrence Berkeley National Laboratory published an article in the Journal of the Audio Engineering Society [PDF] describing a method for digitally imaging the grooves of a record and then processing the image to re-create the sound. The two researchers prototyped it on a 1950s copy of “Goodnight Irene” by The Weavers, giving rise to their system’s name, IRENE (for Image, Reconstruct, Erase Noise, Etc.). Then they tested IRENE on a 1920s Edison wax cylinder [PDF], thereby proving that they had developed a noninvasive technique that could recapture sound from early original recordings.

Curator Carlene Stephens at the National Museum of American History took notice. Could the IRENE team recover sound from the Volta recordings? Working with Peter Alyea, a digital conversion specialist at the Library of Congress, as well as Haber and fellow Berkeley scientist Earl Cornell, Stephens selected six Volta recordings deemed the most likely to reveal sound.

……..

Most of the recovery work was done in Alyea’s lab at the Library of Congress, with the Smithsonian’s Stephens and Shari Stout handling the objects. Stephens also took two recordings—an Edison tinfoil record and a Volta lab graphophone disc from 1881—to Lawrence Berkeley for scanning because the system there had a larger stage. After the initial scan of a recording, custom software analyzed the data to create a sound file. The end result opens our ears to the sounds of the 19th century. (In this excellent video, Stephens and Haber demonstrate IRENE at work.)

………

……..

Viewing 2 posts - 1 through 2 (of 2 total)
Viewing 2 posts - 1 through 2 (of 2 total)

You must be logged in to reply to this topic.