Extracting Audio from Pictures | Hacker News Reader

Extracting Audio from Pictures

mediapreservation.wordpress.com

69 pointsshawndumas10 comments

Few weeks ago I saw a TED video about extracting sound from video through vibrations and figuring out their movement abilities:

https://www.youtube.com/watch?v=npNYP2vzaPo

And now sound from paper/pictures.

A 1860 recording extracted from paper http://www.firstsounds.org/sounds/scott.php

Related: extracting audio from scanning gramophones: http://www.cs.huji.ac.il/~springer/DigitalNeedle/index.html

This was possible only because of the high resolution and optical reproduction techniques they used at the time.

I doubt you'd be able to do the same with a picture of a vinyl record on a magazine page that was taken with a digital camera.

Pretty amazing. This old reddit thread mentions work being done on extracting ambient vibrations potentially imprinted into pottery.

http://www.reddit.com/r/history/comments/1bwyf1/extracting_a...

Sadly I believe this to be a hoax [0] but has been used several times as an interesting plot device - notably in G Benford's short story Time Shards (which I have not read). Similarly it has been proposed that stone stores electrical mental impressions [1] - again the subject of a quite wonderful TV play written by Andrew Nigel Kneale [1]. The most interesting version of the pottery-as-recording story is the one which proposes that Jesus's voice was recorded onto pottery - once in the X Files [0] and, to my knowledge, once in a BBC radio play, where Jesus's final recorded statement is the major plot point (sadly I can't find a reference to the play - but it could indeed be based on the short story of G Benford).

Sorry to go on.

[0] http://itre.cis.upenn.edu/~myl/languagelog/archives/002875.h...

[1] http://en.wikipedia.org/wiki/Stone_Tape

IIRC the xfiles episode was a parody, the "lazurus pot" wasn't real. And the recovered audio of Jesus was him saying "I am the walrus.".

A hoax, sure, but as http://metamodern.com/2014/08/08/recovering-ancient-voices-f... points out it's plausible for there to be some interesting data to recover this way. I haven't heard of any actual attempts, unless you count the Woodbridge letter.

There's a gem right at the end from 1878:

"Oldest known publication of a recording of recognizable phrases in the English language (“Brown University”; “How do you do?”)"

I'm loving the work going on in this space. It's very interesting. My only reservation is it's long-term implications for surveillance.

Reply on news.ycombinator.com