Edison’s phonograph claim faces a 1780s rival

1780s speaking – New findings presented by audio historian Patrick Feaster suggest that a German “speaking machine” exhibited in the 1780s—linked to Georg Theodor Jacob Müller—may have come before Thomas Edison’s 1877 phonograph. The work is not proof, but it reshapes how far

On December 7, 1877, Thomas Edison walked into the offices of Scientific American in New York City and set a metal device on a desk. He turned a crank. Around a dozen staffers gathered close as the machine began to speak in his voice—“Good morning,” it said, “How do you do?”

In the December 22, 1877 issue, Scientific American’s editors described what they had just witnessed. “There can be no doubt. ” they wrote. “but that the inflections are those of nothing else than the human voice.” They published a detailed sketch of the device Edison called a phonograph. and the demonstration landed like a spark in the public imagination.

Virtually overnight, the article catapulted Edison to fame and established the phonograph as the first machine to record and reproduce human speech.

But on May 15, 2026, at the annual meeting of the Association for Recorded Sound Collections in Memphis, audio historian Patrick Feaster asked a different question: what if Edison wasn’t first?

Feaster’s pursuit began more than 20 years earlier. when he encountered a German article from the early 1900s surveying mechanical devices that synthesized some sounds of human speech—without recording them. The article pointed to a man identified only by his last name. Müller. who had exhibited a talking machine in the 1780s. The author called the earlier machine an obvious hoax, but Feaster was intrigued.

Over the following two decades. his intermittent investigations uncovered more references to Müller and his “speaking machine. ” including a book describing the device from 1788—the same year the machine was exhibited in Erlangen. Germany. Feaster also found two eyewitness accounts that matched on key details.

The speaking machine, as those accounts described it, was about 3.5 feet wide and 2.5 feet high. It was deep and flanked by two life-size human figures—one male and one female. Each figure rested a hand on top of a cabinet equipped with 34 “speech mechanisms” resembling organ pipes. alongside levers. rollers. cylinders. clockwork mechanisms and 10 bellows.

Still, other accounts described Müller’s device differently, as a puppet that conversed with audiences.

Then Feaster made a discovery in January that pulled the story into sharper focus: there were two Müllers. Both demonstrated speaking machines in Germany in the 1780s, and neither device has survived.

One was Laurentius Müller, and his speaking puppet was documented as a hoax. The other was Georg Theodor Jacob Müller, a devotee of medicine and mechanical sciences. Feaster says he was struck by contemporaneous accounts of Georg Theodor Jacob Müller’s machine. including a testimonial by physicist Johann Tobias Mayer.

In Mayer’s account. sounds moved from the top of the machine through tubes—carrying vibrations up through the arms of the two figures and into their mouths—producing distinct male and female voices. Mayer. Feaster says. cautioned that “No one will be convinced that the human voice has been achieved perfectly. ” but argued the result could be clearer in a certain test: if the figures were removed and listeners pressed their ears directly on the hole at the top of the cabinet. the speech became clearer.

The machine’s repertoire, according to those records, went beyond simple imitation. It included answers to 12 riddles. passages from books. and laughing. crying and kissing sounds. as well as arias sung in both male and female voices—accomplishments Edison’s phonograph would later perform by recording and playing back human speech.

And yet Mayer took the device for granted as a fake in the same way many contemporaries did. “Everyone assumed that no machine could really do what Müller’s was supposedly doing,” Feaster says.

Two features, however, are what Feaster now leans on as the strongest cracks in the dismissal.

First, Müller reportedly mentioned an artificial ear. That mechanism simulated the human eardrum, gathered sound from the air, and was used during the 1780s as a hearing aid. Feaster argues an artificial ear could have been part of a recording device.

Second was an echo-like behavior. When audience members spoke three or four words into the ear of one of the figures. they heard those same three or four words again in their own voices after a delay. A natural echo with a delay long enough for those words to be enunciated clearly. Feaster notes. would require a much bigger volume of space than the interior of Müller’s cabinet. So if the repeated words weren’t simply an echo. Müller might have used some kind of mechanical technology to record and play them back.

Jacob Smith, a media historian at Northwestern University, frames Feaster’s contribution a bit differently. Even if Müller was a fraud, Smith says, “Patrick [Feaster] has given us a richer picture of the horizon of imagination surrounding talking machines long before Edison.”

Feaster’s work has already changed how researchers talk about earlier speech technologies. In 2008. he and several of his sound historian colleagues established that an invention of the late 1850s was likely the first to capture sounds on paper. That device. the phonautograph. was invented by a French typesetter and channeled sound vibrations from an artificial ear to a stylus that transcribed those vibrations on soot-coated paper in the form of seismographlike tracings. Feaster and his collaborators even used digital technology to turn those soot tracings into an audio recording—breaths that once actually passed through human lips.

Smith also points to the phonograph itself as a reminder of how late surprises can be. “In my class on the history of recorded sound,” Smith says, “the students are always surprised at how ‘low tech’ [the phonograph] is and that, technically speaking, it could have been invented much earlier.”

Maybe it was.

For now, Feaster says the evidence that Georg Theodor Jacob Müller created some version of a phonograph remains “intriguing, inconclusive—and elusive.” That hasn’t slowed him down. This week, he is in Germany hunting for more clues.

The story. at its core. is simple: Edison’s 1877 demonstration and Scientific American’s 1877 description remain central to the history of recorded speech. But Feaster’s case forces the timeline to wobble—because if the cabinet in Erlangen held more than a puppet’s tricks. it may mean the urge to trap the human voice in gears and mechanisms began far earlier than anyone expected.

Edison phonograph speaking machine Patrick Feaster Association for Recorded Sound Collections artificial ear phonautograph media history mechanical devices recorded sound