Freesound Forums

They've discovered a French recording of sound that predates Edison's.

http://news.yahoo.com/s/ap/20080328/ap_on_hi_te/earliest_recording

The tune was captured using a phonautograph, a device created by Parisian inventor Edouard-Leon Scott de Martinville that created visual recordings of sound waves. Using a needle that moved in response to sound, the phonautograph etched sound waves into paper coated with soot from an oil lamp. "When I first heard the recording as you hear it ... it was magical, so ethereal," Giovannoni said. "The fact is it's recorded in smoke. The voice is coming out from behind this screen of aural smoke."

Does anyone know where to find an audio file of this?!

Found it -

http://graphics8.nytimes.com/audiosrc/arts/1860v2.mp3

And here is a text about a very cool hoax -

http://itre.cis.upenn.edu/~myl/languagelog/archives/002875.html

Belgian researchers have been able to use computer scans of the grooves in 6,500-year-old pottery to extract sounds -- including talking and laughter -- made by the vibrations of the tools used to make the pottery.

If only it were true. Some fiction should be true.

Very cool performance, HK. Some futurist influence? You should upload an excerpt here - inaudibility is no excuse! The NYT link is for real, and here's a BBC story, including an interview with the lead scientist who made the sound come alive.

http://news.bbc.co.uk/player/nol/newsid_7310000/newsid_7318200/7318205.stm?bw=bb&mp=rm&asb=1&news=1&bbcws=1

I think this story is a gem. Also, in the text link I posted above there is the following passage by Charles Babbage, the famous 19th century computer nerd -

Thus considered, what a strange chaos is this wide atmosphere we breathe! Every atom, impressed with good and with ill, retains at once the motions which philosophers and sages have imparted to it, mixed and combined in ten thousand ways with all that is worthless and base. The air itself is one vast library, on whose pages are for ever written all that man has ever said or woman whispered.

thanos
And here is a text about a very cool hoax -

http://itre.cis.upenn.edu/~myl/languagelog/archives/002875.html

Belgian researchers have been able to use computer scans of the grooves in 6,500-year-old pottery to extract sounds -- including talking and laughter -- made by the vibrations of the tools used to make the pottery.

If only it were true. Some fiction should be true.

in my opinion, something like this is possible...

BUT, :evil: , for that case,
someone say that it’s a 2005 April’s Fools joke,
check the peoples comments:
http://liftlab.com/think/nova/2006/02/19/sound-and-ceramics-6500-yo-voices-recorded-in-pottery/

I search a little, and found theses links:

- http://graphics8.nytimes.com/audiosrc/arts/1860.mp3
- http://www.firstsounds.org/sounds/

hammerklavier
thanos,
terrific post!

Years back, we did this "avant garde" piece where we brought out archaeological scholars (real), had them take out some ruins (not real) and---with instruments, scalpels and tons of scholarly whispers---drew out a piece of stone that contained the "original sound-wave patterns from an earthquake that wiped out the culture". They then translated the waves into "synthesizer" waves---fake!---and played them back: Well, we'd recorded it earlier---it was this ethereal, crackly rumble-and-crash track, with screams and crashes, made on an old Wollensak reel-to-reel! (It was cheap!) The audience gasped and moaned, and was amazed. Then we told them it was fake & said, "get angry: We'll record you and include it in the next performance". So they stood up & booed & bellowed (& broke down laughing), and we made that into another 'ancient' soundtrack for the next show, and so it went. (This is the only way we got audiences!) It went on about 6 shows, until no one came---and that was the end of the work. I still have the reel to reel, and it's inaudible.

A great thought-inducing post. Anymore---please share!

(ps---is the NYT real? If it is, it's just chilling!)

LMAO, awesome.

HardPCM, that Firstsounds link is great! There are several mp3s of the phonoautograph. The sounds are so evocative, it's like spirits trapped in another dimension and trying to get out.

What are the rules for uploading these to Freesound? They are Creative Commons licensed.

Wow. That clip of the girl singing "Au Clair de la Lune" is just Amazing. it sounds like some one trying to reach out to you from the other side or something. I wish I could find a way to get that same feel for some of the stuff I'm doing... Amazing post thanos! Just curious though what exactly were you searching for when you found this gem?

On an unrelated side note I'm working double time on the time machine after hearing about those concerts HK. lol You'll know it was successful when you Suddenly remember 4 people in the audience.

HK - For the longest time process never really interested me, it was always the result that focused my attention. Nowadays I work believing the process is as valuable as the result and often more so. Sometimes I start with a method that leads to an idea, sometimes I start with an idea that leads to a method, one determines the other. I feel that's somehow connected with the cool performance piece you describe...? There are a zillion ways of making something good, right now process counts more for me (that's it - your piece was all about process, the result was totally unpredictable). And this is why the phonoautograph strikes a real chord - that garbled voice singing clair de la lune could probably be mimicked and even improved by a Russian glitch DJ, but I wouldn't love it. Would you? The fact that it was recorded on paper - covered in soot from an oil lamp - makes it something I want to reach out for, knowing how it was made turns it into something warm and alive.

Lysander - Glad you heard what I heard! Somehow this story made it to the front page on the BBC News site, probably the Beeb engineers flipped out over it. Obviously, if you want to recreate that sound you better get some paper and an oil lamp...

thanos- making a phonoautograph seems entirely possible actually after looking at some diagrams from the century dictionary. I wish I could do it but I've got a feeling my wife won't allow it as I should probably be job hunting instead of tinkering. Here is an enlarged picture of the diagram from the century dictionary. If anyone can figure out what that word ' and movable piece at a, by which the position of the _____ points can be regulated ;' is Please let me know, as I'm trying to make a restored copy of this Diagram and I honestly don't know what kind of points it's regulating
http://img291.imageshack.us/img291/3954/39392148dt3.th.png

Lysander, if you can replicate that contraption, I will make all the excuses necessary to your wife! What a project. The tricky part might be decoding the paper afterward because the phonoautograph was intended to visualize sounds, not reproduce them - the people at Firstsounds developed some kind of method to scan the scratches and transform it back to audio. Kind of like how Audacity can transform jpgs to sound files? Hm. Nah. Very different. Although if you feed Audacity images of geometric shapes it does odd things. I wonder what it would do with a high contrast image of that phonoautogram?

The word is 'modal' - isn't this just a lever for adjusting the needle on the paper?

Hammerklavier, this performance of yours is a riot! It challenges and makes people laugh. And call the police. Perfect night out at the theatre. (I had a less than perfect night out at the theatre yesterday, if only they'd stopped the show and said what you said...) It's so cool for an audience to intervene, to be invited by the cast. The open unpredictability of it. If I found a group that could do that well I'd go every Sunday, theatre as church of the absurd. You still do this sort of thing? Maybe at the first annual Freesound Festival you could share a tent with Bram's massive recording people.

In the BBC interview, they said the focus of Firstsounds is on early sound recordings, cylinders and early LPs and such. The project's goal is to 'read' the audio optically so they don't touch or damage the antique materials. But I think the research could easily go in other directions - if there is the slightest evidence of sound recorded somehow on unlikely objects, what a new world it would open. It's all about the process, of course, engineering the right instruments to read the right marks.

(Incidentally, one of the Firstsounds gang - Patrick Feaster - is at IU).

lysander darkstar
thanos- making a phonoautograph seems entirely possible actually after looking at some diagrams from the century dictionary. I wish I could do it but I've got a feeling my wife won't allow it as I should probably be job hunting instead of tinkering. Here is an enlarged picture of the diagram from the century dictionary. If anyone can figure out what that word ' and movable piece at a, by which the position of the _____ points can be regulated ;' is Please let me know, as I'm trying to make a restored copy of this Diagram and I honestly don't know what kind of points it's regulating
http://img291.imageshack.us/img291/3954/39392148dt3.th.png

the description of the invention is here:
- http://en.wikipedia.org/wiki/Phonograph#The_phonautograph

What I understand of that:
is that the cylinder will concentrate the physical vibrations,
into the metal key and more the pitch/frequency/cycles will be high,
more dots will be drawn for the same time/duration/lenght on the paper,
hope it was usefull...

You must see that:

- http://www.talkingmachine.org/phonautograph.html

I described the bad thing,
In fact the thing just draw an regular sound wave representation hock:

Yes, the machine looks pretty straightforward. Beautiful, too. I think it could make really nice lines on paper.

Back in the digital world, here is an absurd experiment with a crap waveform of Au Clair de la Lune -

http://thanosthanosthanosthanos.googlepages.com/auclairdelalune

Thanos- Link in your last post: The sounds didn't work for me (methinks it was the damned player *pouts* and I was hoping to hear that too *sigh*)

I honestly think this thing (for the most part) is reproducible. I only wish I had the time/tools... I think I'm going to make it into a hobby... I suggested the idea to her and she rolled her eyes.. lol.

and as far as playing it back I'm at a loss I honestly Don't have a clue at that point BUT I'm sure that the guys at First sound (if I told them what I was doing) would be more than happy to share their program with me...

The only hard part seems like that little arm(thing) the needle is resting on... I'm not Quite sure on that but honestly the device itself seems REALLY easy.

It'll probably go into my pile of pipe dreams (It's been growing exponentially since I got married) ((and I'd wager it will continue its rampant growth until I retire)). But I will say this much I'll keep it on the Top of said pile. Because this is something I'd REALLY like to have.

@HK- Man, I don't see why you'd have trouble pulling an audience with a show like that. I'd love to have been there... Maybe you were just too early, I'm sure it'd go over well nowadays.

HK - On an intuitive level, I think you're absolutely right about the relation of structures. Even if the science can't explain everything yet, it's a fact that we humans can instinctively sense a quality from one form represented in another - a waveform of Bach has strong balance and visual harmony that isn't purely about maths, for example. The quality of the phonoautogram's sound evokes a sense of time and dimension which I think a simple digital distortion couldn't, but I don't know if there can be any good explanation. Your description of capturing inherent structures sounds a lot like homeopathic medicine, where a chemical is diluted a zillion times with water and supposedly still manages to transmit its properties (I don't believe in that stuff because I've seen it not work, but I've also seen Picasso's forms diluted and transformed while still maintaining something from the master's hand. Maybe I was fooling myself of course, but the laws of science don't necessarily apply to art and emotion. Or maybe they do and we just haven't understood how yet.) Process, process...

There are more questions about process - like how exactly did the Firstsounders transform the paper back into sound? A waveform is all about amplitude, right? It doesn't have any information of tone or timbre. So how did a woman's voice appear? How much educated guesswork was involved in filling the gaps of information? How much of this sound was actually recorded in the lines and how much was 'reconstructed'? It may come down to some very subjective judgements on the audio interpretation of those lines. Beautifully executed of course, and maybe even extremely accurate. But I really don't understand how they did it.

Lysander - try using a different browser to see that experiment, maybe it helps. (I've never heard a giant squid fart before, who knew the sound was 'embedded' in the voice of Yvonne Printemps...all these years...just waiting to be released...) This is the first time I've used g**glepages & found that Safari & Firefox read it just fine but Opera won't load the players. Sigh. Someday everything will be perfect. Your phonoautograph project is so cool, it could be a real steampunk masterpiece and it'll definitely make beautiful lines. Maybe tell your wife you want to see her voice? Romantic, no?

thamos
There are more questions about process - like how exactly did the Firstsounders transform the paper back into sound? A waveform is all about amplitude, right? It doesn't have any information of tone or timbre. So how did a woman's voice appear? How much educated guesswork was involved in filling the gaps of information? How much of this sound was actually recorded in the lines and how much was 'reconstructed'? It may come down to some very subjective judgements on the audio interpretation of those lines. Beautifully executed of course, and maybe even extremely accurate. But I really don't understand how they did it.

First, I hate big dialogue, :evil: ...

After,
The numerical rules of numerical sampling can applied to that principle,
else to take an electronic analog signal and numerise it,
we take and visual analog signal and numerise it,
we get an picture so after we process it
to find the level at each pixel horizontal step and that"S all...

Are you saying the phonoautograph's waveform does contain tone information? How? Can you illustrate that for me?

Mr., you need to learn a little about the fundamentals of music or pitch:
http://en.wikipedia.org/wiki/Pitch_%28music%29

Find one of these waveform picture with an good pixel resolution and I will give you the demonstration