Skip to main content

Hi folks. This seemed the best sub forum for this, please forgive and advise if i'm wrong.

A friend of mine passed away about a year ago, young kid, really sad. I found this audio recording I made in which he's talking for about 20 minutes straight (he liked to talk--most 17 year olds do!)

well about 10 minutes in the audi becomes really garbled, scrambled, I would call it. you can hear that it's a regular patter, because you can hear that much. I feel like there must be some way to run this through some fancy software and put it back together. It'd mean a lot to me if anyone had any ideas.

here's a sample of the garbled audio. feel free to download and experiment.

(Dead Link Removed)

Thanks all!

Topic Tags

Comments

anonymous Sat, 05/10/2014 - 04:02

When you found this audio, what was the recording on? - I mean, was it a cassette tape, a DAT tape, a CD?

A cursory listen gives the impression that this section had been transferred from another source (digital) with a "fast cue" function... this function allowed the user the ability to "scroll" through certain sections in a fast wind "cue" mode. It was helpful in finding sections of tape that had silence, and you could fast cue, listening to the section of tape to find where audio would appear. It was also used for finding the beginnings and ends of audio content.

This sample sounds to me as if it was digital at some point and that someone, during the transfer, engaged this fast cue function. The problem with that is that it didn't fast-wind in a linear fashion...unlike regular analog tape, which would present the audio in the same context - just much quicker - instead it used digital samples, and when those samples are sped up in that mode, they can get choppy and altered because very often, data is left out during this fast cue mode.

Unless there is some unique audio forensic software available that I'm unaware of, unfortunately, I don't think that there is much you can do to retrieve this data.

I could be wrong though... you should wait for others to chime in.

paulears Sat, 05/10/2014 - 05:48

I had a DAT machine that did that quite a lot - and was characteristic of when the data correction rate was up in the thousands, rather than the usual less than 10 range. The error correction is constantly filling up the buffer then jumping, then doing it again, so you end up with a complete data stream composed of fragments and gaps. As far as I'm aware, there isn't any recovery possible because the data didn't get recorded - as it was going to tape or disc bits didn't get recorded and the error correction is allowing playback, but in this wrecked format. Most machines would mute the audio before it gets this bad. You can reproduce the effect on a variable speed DAT recorder by using 2 or 3 times speed playback if your machine has a shuttle knob - it sounds exactly the same as is kind of ok for scrubbing, where the disjointed effect at original pitch is better than analog going through like Mickey Mouse.

I've actually got lots of old recordings that can't be replayed now that do this to a similar degree. I did have one MD machine - an early one that would make this noise on dodgy discs (A Tascam 801) but most now mute.

What medium is it on? It's rare on CDs, but common on old digital tape.

fearfeasog Sun, 05/11/2014 - 16:29

hey guys thanks for replying.

This was recorded on a small Sony IC voice recorder (from at least 5 years ago), directly to mp3. It pretty much came off the recorder sounding like this, transferred as an mp3 file directly to my computer. Totally bummed. I was afraid that there'd be missing info as you said paulears, and DonnyThompson I did the same thing! you can hear the pattern though, right. bits repeat maybe a second after the original, but it's all garble together nonetheless. aargh. I guess I should just consider it "not ever recorded" like so many other moments we had with Jon. Thought there was a chance though. :)

Mark

audiokid Sun, 05/11/2014 - 16:57

It sounds like the clock isn't grabbing the DA. I've had that happen but it was connected to the interface, just at the wrong sample rate. I don't think, as Donny tried, to slow it down because its already been captured out of sync. I'm only guessing but you would have to start from the Sony again. When you captured this to your computer, how did you do that? Through usb or some digital transfer or actual AD via a sound card or converter?

fearfeasog Sun, 05/11/2014 - 18:24

audiokid, post: 414690, member: 1 wrote: It sounds like the clock isn't grabbing the DA. I've had that happen but it was connected to the interface, just at the wrong sample rate. I don't think, as Donny tried, to slow it down because its already been captured out of sync. I'm only guessing but you would have to start from the Sony again. When you captured this to your computer, how did you do that? Through usb or some digital transfer or actual AD via a sound card or converter?

through USB. the recorder has a built in USB plug under a protective cover. I figure the damage happened during recording, probably.

audiokid Sun, 05/11/2014 - 18:46

I'm only guessing here, and have no idea about your recorder but from the sound of the audio posted, I'd be more inclined to say its the usb interface that caused this for you. You didn't lock the computer to the recorder, yet it transferred the audio.
Are you able to hear the original recording without it connected to anything?

RemyRAD Sun, 05/11/2014 - 22:13

The possibility exists that the bit depth, sample rate are not matching or Mono versus stereo? Maybe if the solid-state Sony recorder is sent back to Sony? Especially with what you indicated to us, what this is all about. It couldn't hurt to ask Sony? They might find it in their heart to try and recover it? They are a most remarkable company after all. This would certainly mean something to his family.

I tried to get into forensic audio but unless you are a coder genius? The analog days are over. Which was my strong point. This is most unfortunate. I also tried to make some sense of it but no. It's totally randomized. It's algorithmically clobbered. It's kind of like your late 17-year-old friend. How prophetic.

What a bummer.........

fearfeasog Mon, 05/12/2014 - 07:12

audiokid, post: 414693, member: 1 wrote: I'm only guessing here, and have no idea about your recorder but from the sound of the audio posted, I'd be more inclined to say its the usb interface that caused this for you. You didn't lock the computer to the recorder, yet it transferred the audio.
Are you able to hear the original recording without it connected to anything?

the original is no longer on the recorder, unfortunately. :(

RemyRAD, post: 414694, member: 26269 wrote: The possibility exists that the bit depth, sample rate are not matching or Mono versus stereo? Maybe if the solid-state Sony recorder is sent back to Sony? Especially with what you indicated to us, what this is all about. It couldn't hurt to ask Sony? They might find it in their heart to try and recover it? They are a most remarkable company after all. This would certainly mean something to his family.

I tried to get into forensic audio but unless you are a coder genius? The analog days are over. Which was my strong point. This is most unfortunate. I also tried to make some sense of it but no. It's totally randomized. It's algorithmically clobbered. It's kind of like your late 17-year-old friend. How prophetic.

What a bummer.........

yep, kindof a bummer. thanks for the advice.