Forgot your password?
typodupeerror
Music Media Technology

Making Old Sound Recordings Audible Again 172

Posted by samzenpus
from the does-it-work-on-8-track dept.
orgelspieler writes "NPR is running a story on a safe way to reproduce sound from ancient phonographs that would otherwise be unplayable. The system, called IRENE, was installed in the Library of Congress last year. It can be used to replay records that are scratched, worn, broken, or just too fragile to play with a needle. It scans the groves optically and processes them into a sound file at speeds approaching real time. IRENE is great at removing pops and skips, but can add some hiss. Researchers are also working on a 3D model that is better at removing hiss."
This discussion has been archived. No new comments can be posted.

Making Old Sound Recordings Audible Again

Comments Filter:
  • by Anonymous Coward on Wednesday July 18, 2007 @07:55PM (#19909091)
    Oh wait, never mind...

    (but i swear that's what my mind picked up initially!!)
  • I remember a few years back some guy had some way of using a laser to play the needle; is this the same thing improved?
    • There was a Slashdot story a few years back about a guy who had written software that let him pop a record on his scanner and play it. It scanned the record and then reconstructed the grooves from the image. The sound quality was terrible, but it was a nice proof-of-concept.

      This is the same thing, but with a much better scanner.

    • by nurb432 (527695)
      You can buy optical turntables. They aren't cheap but its not just 'some guy'.

      This sounds more like the old joke about photocopying your records to 'copy them'.
  • Can anyone find a link to source code for this?
    • by Tribbin (565963)
      That is what they try to recover here; the analog source code information of the audio on the broken phonograph.

      I was not able to find any transcodings of the audio, to answer your question.
  • IRENE is great at removing pops and skips, but can add some hiss

    Can add some hiss to what? To the perfect Hi-Fi quality you are expected to get out of a century old phonograph?
    • Re: (Score:2, Informative)

      by og_shift8 (949137)
      looks like 3d noise reduction is done in the image domain - not in the audio.
    • by Ungrounded Lightning (62228) on Wednesday July 18, 2007 @08:13PM (#19909237) Journal
      Can add some hiss to what? To the perfect Hi-Fi quality you are expected to get out of a century old phonograph?

      To the level of his that the recording itself actually contains.

      Old recordings actually did a very good job of making a record of the actual sound. But dust on and damage to the surface produced artifacts in the output signal when played with a needle.

      Optical techniques can identify the actual flat surface of the groove and ignore the artifacts. But digital approaches to performing this scan and/or encoding the result add errors from quantization and digitizer nonlinearity, which appears as added hiss - the amount depending on the resolution and quality of the converter and/or scanner.
      • Re: (Score:3, Interesting)

        by schwaang (667808)
        Interesting. I didn't RTFA, but my first thought was that the optical technique was picking up hiss (high frequency) that existed on the originally produced media, but that was smoothed out (i.e. not reproduced) by the mechanical arm-and-needle.
        • by arodland (127775)
          Nah, that would be dead simple to correct for. Just create a lowpass filter that matches the response of a really good physical turntable. If the sampled data is accurate, then that would be all you would have to run it through to get beautiful reproduction. But if the sampling process has errors, you're slightly more screwed :)
          • The way I do my LPs (Score:3, Informative)

            by Pope (17780)
            The way I do my vinyl restoration is using Cool Edit Pro aka Adobe Audition. For the noise reduction pass, what you do is take a sample of the between-song space, which is normally empty, and then mathematically subtract this 'silence' from the entire sound file. Voila! The surface noise is eliminated from you recording, and you can then do any more de-clicking, etc. as you will.
        • by adolf (21054)
          You're thinking too much :)

          If there's no sound with which to move the needle about, then the needle doesn't move. If the needle doesn't move, then the groove doesn't change. And if the groove doesn't change, then the playback device (ideally) produces no sound.

          There's simply no mechanism for hiss, as we know it from electronic recordings and instrumentation, to enter the picture.

          • by schwaang (667808)
            Yeah but what about error introduced by the mass production technique? I'm saying, a faithfully recorded zero-energy ("perfectly" smooth -> no sound) groove from 1910, when put under a 21st century scanning tunneling microscope, looks pretty darn craggy, in a random high-frequency ("hiss") kind of way. You wouldn't want to try playing back the image of the groove without taking into account what a 1910 needle would "see".

            I'm not saying I'm right about the hiss existing in production though. Just makin
            • by adolf (21054)
              True, to a point.

              But it still wouldn't show up as hiss, per se: If, on the duplicating machine, the master has zero groove variation, then the copy will as well.

              Things are sure to get a lot more interesting when there's something other than silence happening, but it will probably manifest itself in the form of harmonic or intermodulation distortion, neither of which are generally all that random (being products of the original signal).

    • by Tribbin (565963)
      There is probably just noise in the depth measurement. That is technical information-loss; even if the source is of bad quality.
    • ... if all the pop is removed?
    • the perfect Hi-Fi quality you are expected to get out of a century old phonograph?

      Surprise, surprise, listen to the fine samples. The first collection sounds like it was recorded yesterday. The technique is unbelievably excellent. This is very good news for music preservation.

      • Awesome! I love this kind of technology. A) Its good to see a generations information going into the void, B) its awesome to see new ways to collect data about real world objects (like records) for future generations.
      • by Wolfrider (856)
        --They could do something similar for preserving obsolete data formats.

        o Convert the existing data to UTF8 -> binary.

        o Find a rock sheet (thin enough to be fairly light, but not thin enough to break easily.)

        o Permanently etch the binary into the rock sheet with a frickin' laser beam. (Wielded by a shark of course.)

        o Store the rock in a cool, dry place. Preferably out of sight, and armored against the inevitable barbarian/Zombie hordes.

        o ???

        o Profit!

        --You heard it here 1st.
        • We're the RIAA and we'll just relieve you of that.

          All money is OURS! Bwahahaha...

          We won't stop until we can collect on a bee's fart in the forest. Bwahahaha...

          And if you don't like it, well you can just take this here ice pick and shove into your ears. Bwahahaha...
      • by PopeRatzo (965947) *
        You think they can do something with my old copy of The Velvet Underground? I think there's peanut butter stuck in some of the grooves. I rolled so many joints on the cover that it smells like a Columbian's armpit. But the record has great sentimental value. I bought that bitch in 1972 when I was a sophomore in HS and it left a permanent imprint on my psyche. Somehow, I still have it in my basement.
      • Re: (Score:2, Informative)

        by GetAssista (1130195)
        [i]The first collection sounds like it was recorded yesterday. The technique is unbelievably excellent. This is very good news for music preservation.[/i] There is no way you can capture higher frequences material from old LP, not to mention phonograph. It's just not there and hence can't be picked up whatever the device. LP's from 40's which I work with, have records go up to 6-8kHz, and nothing higher except hiss. That's a lot. Try it yourself with this low-pass filter applied to modern recording and see
    • by Tony Hoyle (11698)
      It actually adds a lot of hiss - the IRENE versions of the songs sounds *much* worse than the originals because the occasional crackle is replaced by a high level of background hiss - in the second sample almost drowning out the recording.

      Interesting experiment but until they can produce a recording at least as good as a standard (or laser) player then it's not a useful technique.
      • If they are looking directly at the physical groove then I wonder where the hiss is coming from. Is it inherent in the grooves and does a needle create a natural dampening effect when played normally?

        I agree they shouldn't spend millions on a mass digitization effort until they improve the audio quality.
  • by Anonymous Coward
    these audio recordings aren't degraded, they just have some sort of ancient DRM on them. many years from now they'll find some ancient music cds of ours with DRM on them and think they are degraded too.
  • by Anonymous Coward
    how fidel do you want to be?

    OK, so the real quote is "Speed costs money, how fast do you want to go?" and is usually applied to hot rods. The analogy to sound is pretty accurate though.

    At some point, you can't just pop a disk or cylinder into a machine and have everything automatic. Expensive people have to get involved. In theory, as long as the signal is there, you can re-construct it in the face of a huge amount of noise. The process is not dissimilar to getting the data off a trashed hard drive. In
  • I've been a fan of ancient public domain music for a while now. I hope they are kind enough to post these on a website for our listening pleasure.
    • In 2067 (Score:3, Informative)

      by tepples (727027)

      I've been a fan of ancient public domain music for a while now. I hope they are kind enough to post these on a website for our listening pleasure.
      In 1972, when the U.S. Congress phased out state law copyright in sound recordings, Congress allowed these copyrights to continue for one full federal copyright term. This term ends in 2067.
      • It is indeed a very sad state of affairs in the US with regards to recordings. I was shocked when I learned of it. But perhaps the original poster is from other shores.
    • by ultranova (717540)

      I've been a fan of ancient public domain music for a while now. I hope they are kind enough to post these on a website for our listening pleasure.

      Public domain ? These recordings are barely a century old, they are hardly in public domain yet. In fact I expect the people who invented this device to be sued for bypassing an effective copy prevention device; after all, these things are not that dissimilar to the limited-time degrading DVDs, now are they ?

      • Re: (Score:3, Informative)

        by j-cloth (862412)
        Except in intent. The century old recordings were not purposefully put on media ment to degrade. Limited time DVDs are.
  • I bet a lot of floridian orange growers do a lot of "grove scanning" as well.
  • NPR on /., again? (Score:3, Insightful)

    by aethera (248722) on Wednesday July 18, 2007 @08:19PM (#19909283)
    Now I know that my local NPR station [wuky.org] is skewing towards a younger market. Heck, they dropped the classical years ago and play a decent mix of non-mainstream, non-corporate, though I wouldn't go so far as to say indie rock during the day. But when I started reading on slashdot regularly ten ( 10!) years ago I would have never expected the relatively common recurrence of NPR articles making the front page. Are we all getting that old, or am I just getting old enough to notice it?

    No offense to some of the bright high school students and undergrads who comment here...you're appreciated, sometimes for you're youthful naivety, but appreciated nonetheless.

    • by Lumpy (12016) on Wednesday July 18, 2007 @09:13PM (#19909629) Homepage
      Dude, I was enjoying some Chemical Brothers on NPR last sunday. I though I tuned to the MSU student radio station but noticed that I was on the Statewide NPR station (they transmit on 4 different frequencies at incredibly high power to cover almost all of lower michigan).

      They also played some newer Information Society and then finished with some DonJuan Dracula before they broke.

      I was freaked to hear some really progressive music played on NPR. They either must be desperate to attract new listeners or don't care they will turn off the old farts who grimace at hearing that "pounding hippy music"

      • Re: (Score:3, Insightful)

        by bdjacobson (1094909)

        Dude, I was enjoying some Chemical Brothers on NPR last sunday. I though I tuned to the MSU student radio station but noticed that I was on the Statewide NPR station (they transmit on 4 different frequencies at incredibly high power to cover almost all of lower michigan).

        They also played some newer Information Society and then finished with some DonJuan Dracula before they broke.

        I was freaked to hear some really progressive music played on NPR. They either must be desperate to attract new listeners or don't care they will turn off the old farts who grimace at hearing that "pounding hippy music"

        I applaud them for this.

        There's a time when you stop listening to music to feel and start listening for entertainment. At this same point, you realize most of the MTV music sucks.

        When your motivation for listening to the music is entertainment, I would define that as simply searching for something new...a new outlook on the old chord progressions, if you will. Or out of the ordinary chord progressions, etc.

        Hence again NPR caters to the intellectual type. First they did it with Classical music, now they do

        • I think it's partially a generational thing. Here in the UK, our national general music station, BBC Radio 2, long had a reputation of playing sixties music by day and jazz and big band music by night. Its target audience was generally the over 40s. These days it has pulled its audience back to the over 30s and has actually paid attention to what the over 30s listen to, and has become the best station in the country. But then again, I'm well over 30.
      • by megabeck42 (45659)
        U of M's Michigan Radio or WCMU from Central?
    • Yes.

      Next question?

      Get off my lawn! Damn kids.

  • Hmm... (Score:5, Funny)

    by Masato (567927) on Wednesday July 18, 2007 @08:35PM (#19909401) Journal
    I wonder if they can help this guy [youtube.com]?
  • Dupe! (Score:5, Informative)

    by phantomfive (622387) on Wednesday July 18, 2007 @08:38PM (#19909429) Journal
    I don't know what slashdot is coming to, this is a total dupe [slashdot.org]! OK, so that story is from 2005, so what? OK, so I remember slashdot stories from two years ago. So what? Doesn't mean I don't have a life. Right? Right?
    (Actually that other story is pretty cool, has some neat pictures and goes more in depth on the technology. And theres a nice thread [slashdot.org] talking about three-grooved records).
    --
    Looking for a C/C++ job in Silicon Valley? [slashdot.org]
    • I don't know what slashdot is coming to, this is a total dupe! OK, so that story is from 2005, so what?

      That's really similar to what the NPR story described, but AFAICT it's a different project performed by different people and wasn't thrown together in an evening (as the end of the page you linked to says that project was). FWIW.

    • by mzs (595629)
      It is not a dupe. That old article is in a similar vein, but the IRENE system lets archivists read ancient wax cylinders at a rapid pace. Also only recently (in the past year) has there been any progress toward removing the peculiar periodic hiss.
  • by sdo1 (213835) on Wednesday July 18, 2007 @09:18PM (#19909673) Journal
    I'm on the lookout for a system that will make new recordings audible again.

    Virtually every new recording is compressed to the Nth degree with no sense of dynamics and utterly bereft of feeling and life. MP3 compression only makes bad recordings worse.

    -S
    • Try listening to some bands that actually know what dynamics are and incorporate them into their music. The recordings I have of those bands tend not to be compressed (note: dynamic compression, not data compression) because the band wouldn't let the label release it like that. The other interesting thing is that these bands tend not to be on major labels, so you're not supporting the RIAA by buying their music.
    • How exactly does on quantify "feeling" and "life," especially with something as subjective as music? Yes, uncompressed is generally better than compressed, but the music itself is MOST influenced by the subjective tastes of the listener. For example, I would rather listen to a crappy 4th-generation cassette tape copy of Metallica's "Kill 'Em All" than to hear their latest album performed live right in front of me. To me the former would have much more feeling and life than the latter, whatever its technical
  • And they don't just pick up an off-the-shelf laser turntable because...?

    • ... because it was recorded on cylinders?
    • Re: (Score:3, Informative)

      by FranTaylor (164577)
      Because you didn't read the article! Laser record scanners have to spin the record around, this one scans it in place, so you can scan broken records or old Edison cylinders.
  • by jollyreaper (513215) on Wednesday July 18, 2007 @10:46PM (#19910319)
    The digital reproduction just isn't as warm as the analog original we're no longer able to hear in comparison. By the way, have I told you how wonderful my gold-plated connectors are? You can practically hear the money I spent!
    • It was always my understanding that the gold plated connectors are not better because of their fidelity but because they don't corrode. Silver is the best choice for connectivity isn't it but it corrodes fairly easily.
      • by Tony Hoyle (11698)
        Your stereo is full of copper which can corrode equally well.

        If you're running your hifi on a yacht, then it matters, otherwise all the gold stuff is pure snake oil - in fact it makes things worse, introducing resistance between the connections on the stereo.
    • Gold-PLATED? I hope you don't have the gall to actually call yourself an audiophile. No true audiophile would settle for anything less than solid gold connectors. Gold-plated? You might as well just use two cans and a string.
      • Gold-PLATED? I hope you don't have the gall to actually call yourself an audiophile. No true audiophile would settle for anything less than solid gold connectors. Gold-plated? You might as well just use two cans and a string.
        Once my dog barked while I was playing Dark Side of the Moon and I had him put to sleep. Does that make me an audiophile?
  • Cool idea. I hadn't heard about the guy using a scanner to do something like this, so this was a new one to me. If they're successful in the effort to reduce the hiss, this could indeed mean a lot in terms of preserving recordings (as a previous commenter mentioned, and TFA implied).

    Since a lot of people (who obviously didn't RTFA) are confusing this idea with laser turntables, I'm assuming a number of you have experience with them. I am, of course, familiar with the concept, but I've never had the oppo

    • by mshurpik (198339)
      It's a cool idea. I hadn't heard about the guy scanning disks for archival format, so this is new to me. This could mean a lot in terms of preserving recordings.

      A lot of people are experienced with laser turntables. In fact, a recent Spin Magazine survey indicates that laser turntables have overtaken conventional turntables by a ratio of 2:1. By comparison, conventional turntables have a much lower treble range, weaker bass, and higher overall cost. It's the CD generation that is holding music back.
  • Coincidentally, I just spent the last weekend converting some old 78s using a modern (albeit not laser-based) record player. I wrote a little article about it here: http://www.ambor.com/public/78rpm/78rpm.html [ambor.com], including some sample audio clips that show what the raw recording sounds like and then shows what some open source audio restoration software can do.
  • I saw this done years ago at an art exhibit in Seattle at 911 Media Arts. I thought it was cool (in that industrial sense) but I didn't expect to be reading about it as news something like 15 years later.
  • by niktemadur (793971) on Thursday July 19, 2007 @05:48AM (#19912231)
    Even older and of great cultural importance are wax cylinder recordings.

    The old wax cylinder players were also recorders, and they were portable, even if quite bulky. At the turn of the century, explorers from the Royal Geographic Society, for example, were logging these devices around the world, recording songs and rituals of many different peoples, from the folk songs of eastern Europe to war and mating rituals of tribes in the south Pacific.

    These audio documents catalog communities as they were before western industry, politics, etc, seeped in during the course of the twentieth century. Many of the communities recorded in the wax cylinders have probably lost elements of their heritage, if not outright scattered. Think Hawaii, as an example which I don't mean to trivialize, but I'd rather keep it short and simple: old tribal rituals have now become entertainment pandering to the tourists at luaus or at the airport. How about modern hawaiians (or anybody else, for that matter) hearing their ancestors really going at it, psyching themselves up for the hunt at sea, when it was a do-or-die affair?

    Put in another way, I forget who said it (may have been William Burroughs) and I paraphrase: "Once the natives start wearing the t-shirts, that's it, the old magic's gone". And then, there was television... Well, in the wax cylinders, there it is, that old magic.

    One final example: in WFMU, the great radio station from New Jersey, there was a show years ago called The Secret Museum Of The Air, and in a program dedicated to gypsy music, they dug out a recording from 1902, a girl in her village singing a capella to her dead brother, asking him to please visit her in her dreams that night. Even through a century of pops, scratches and hiss, as well as the language barrier, it was an un-fucking-believable, mind blowing thing of extreme poignancy and beauty. Compound that with the very real possibility that nobody alive may sing this song anymore, and it just goes to another, eerie level.

    This stuff needs to be rescued, restored and preserved.
    • Re: (Score:3, Informative)

      by Mordaximus (566304)

      Even older and of great cultural importance are wax cylinder recordings.

      From TFA : "When taking flat photographs, it can create a three-dimensional image of the groove on a record, or on an old wax cylinder. Haber been working with the University of California's Phoebe Hearst Museum of Anthropology, to reconstruct sound from field recordings, like one wax cylinder made around 1911 that features a Native American called Ishi."

      • This is really exciting tech. I'd love to have access to a library of this. It is the closest to a time machine I'll probably ever get.
  • by ehaggis (879721) on Thursday July 19, 2007 @06:59AM (#19912511) Homepage Journal
    The are taking the Library of Congress to court for copying phonographs.
  • I have friends who are DJ's and they've often used laser/light based vinyl decks. Yes this does modelling etc, but it's only really an un-interesting natural extension of it.
    • Wrong. Laser turntables have to spin the record and follow the grooves, producing sound in real-time. This process takes a complete image of the record/cylinder, and processes the image into audio. Since there's no need to spin the record, you can recover sound from records broken in pieces, heavily warped records, etc.

  • So are we talking caveman ancient, or more recent, like maybe the time of the Battle of Thermopylae?
  • Researchers are also working on a 3D model that is better at removing hiss

    A few weeks ago, I was digitiging a record that hasn't made it to CD. For shits & giggles, I ran a noise reduction filter on its lowest setting. Parts of the record sounded like a very low bitrate MP3. What I realized is that sometimes the hiss masks flaws in the recording medium, and removing hiss can sound worse!

  • it would be interesting to load time-synched copies of the same mono recording from a stylus transcription and an optical transcription into a stereo mp3, so you could listen to the differences by playing with the balance.

    i could do it with audacity, but i think it would be useful to have on their web site - i didn't notice such a file there.

Hackers of the world, unite!

Working...