Recorded on the Phonautograph, an invention of Édouard-Léon Scott de
Martinville (Frenchman of Scottish ancestry).
The device used a boar's bristle attached to a vibrating membrane. It
scribed its vibrations onto a surface coated with lampblack.
Scott de Martinville hoped to invent a playback mechanism, but never
did. Fortunately a few of his recordings survived, and in 2008 some
smart fellas in California managed to get a voice from the tracings:
probably Scott de Martinville himself singing (very slowly) the first
few notes of "Au clair de la lune".

Hear it here:

https://www.youtube.com/watch?v=_dbyIDTmHSM

and another version (at wrong speed?) here:

https://www.youtube.com/watch?v=WpXNqdEUhWY

Much more:

https://en.wikipedia.org/wiki/%C3%89douard-L%C3%A9on_Scott_de_Martinville

https://en.wikipedia.org/wiki/Phonautograph

http://www.firstsounds.org/

Ar an deichiú lá de mí Aibreán, scríobh Ross Clark:

> Recorded on the Phonautograph, an invention of Édouard-Léon Scott de
> Martinville (Frenchman of Scottish ancestry).
> The device used a boar's bristle attached to a vibrating membrane. It scribed
> its vibrations onto a surface coated with lampblack.
> Scott de Martinville hoped to invent a playback mechanism, but never
> did. Fortunately a few of his recordings survived, and in 2008 some smart
> fellas in California managed to get a voice from the tracings: probably Scott
> de Martinville himself singing (very slowly) the first few notes of "Au clair
> de la lune".

Subjectively recorded media have slowed regional sound change and perhaps even
reversed local innovations, but I’m not sure how to study cause and effect. The
usual dialect archives do tend to demonstrate more pronounced local features in
the 1950s, which gives effect, but cause is the harder bit.

--
‘As I sat looking up at the Guinness ad, I could never figure out /
How your man stayed up on the surfboard after fourteen pints of stout’
(C. Moore)