Sounds

First Sounds has pioneered the recovery of sounds recorded on phonautograms - many of which were made before Edison invented the phonograph in 1877. The samples below are among the world's earliest sound recordings.

These sounds are licensed under Creative Commons Attribution (by) license and may be redistributed or sampled; all we ask is that you provide First Sounds with a copy of your work. Also, if using the sounds on your website, please save the file and host on your server. Note that these files are not excerpts; they are the full tracks as processed so far.

(The files can be saved by right-clicking on the link and selecting the save option offered by your web browser.)

Au Clair de la Lune--French folk song (1860 Phonautogram)

Scott recorded someone singing an excerpt from the French folksong "Au Clair de la Lune" on April 9, 1860, and deposited the results with the Académie des Sciences in 1861. The existence of a tuning-fork calibration trace allows us to compensate for the irregular recording speed of the hand-cranked cylinder. The sheet contains the beginning line of the second verse - "Au clair de la lune, Pierrot répondit" - and is the earliest audibly recognizable record of the human voice yet recovered.

We released this sound in March 2008, and it has become the icon of humanity's first recording of its own voice. In September we undertook a second restoration using more advanced technologies. Both versions are below.

Au Clair de la Lune--French folk song (mp3, restored September 2008) * NEW *
Au Clair de la Lune--French folk song (mp3, original March 2008 restoration)

Gamme de la Voix - Vocal scale (1860 Phonautogram) * NEW *

On May 17, 1860 - five weeks after recording "Au Clair de la Lune" - Scott and his singer returned to his studio to record a simple vocal scale. He deposited this phonautogram with "Au Clair" and other samples of his work at the Académie des Sciences in 1861. This is the second-oldest recognizable recording of the human voice yet retrieved.

We present two restorations of this recording. The first includes crosstalk from his tuning fork that leaked onto the voice track. The second renders the voice without the tuning fork crossover, as Scott intended to record it.

Gamme de la Voix - Vocal Scale (mp3, with tuning fork crosstalk)
Gamme de la Voix - Vocal Scale (mp3, crosstalk removed)

Phonautographie de la voix humaine à distance--excerpts, at different speeds (1857 Phonautogram)

Scott identified the sheet of phonautograms he deposited with the Institut National de la Propriété Industrielle in 1857 as documenting "the human voice at a distance." Two brief excerpts from two different records on this sheet are the earliest traces of his work played back to date, but his recording methods were not yet sophisticated enough at this time to yield audibly recognizable results. Here we present the two excerpts played at several different speeds.

Phonautographie de la voix humaine à distance--excerpts, at different speeds (mp3)

Diapason at 435 Hz--at sequential stages of restoration (1859 Phonautogram)

Scott attached another phonautogram to the "certificate of addition" he deposited with the Institut National de la Propriété Industrielle in 1859. We believe it to be a record made by a tuning fork vibrating at 435 Hz, then just adopted as the official French reference pitch. This is the oldest recognizable sound yet reproduced and is presented here at successive stages of restoration.

Diapason at 435 Hz--at sequential stages of restoration (mp3)

Metropolitan Elevated Railroad from 40 feet away (1878 Phonautogram)

In 1878, when Thomas Edison was hired to study the objectionable noise produced by the Metropolitan Elevated Railroad in New York City, he turned to the phonautograph, adapting one of his tinfoil phonographs to draw a "readable" lateral waveform. Edison's colleague Charles Batchelor made this particular phonautogram as part of that project in September. We believe the excerpt presented here begins and ends with test shouts, with three specimens of actual train noise in between-the earliest American sounds yet reproduced. Note that pitch fluctuations are due at least in part to the irregular recording speed.