Hello, and thanks for reading...

I absentmindedly returned a rented portable sound recorder before I retrieved the short film's location sound from it, ugh.

The portable sound recorder is an Audio Devices 744T and I had it set up to record track 1 (mic) and track 2 (nothing) to both the internal hard drive and the compact flash. I realized my mistake an hour after I returned it, but they had already formatted the hard drive and the compact flash card. I got the card back in my hands before any more data was written to it.

I tried a handful of recovery software packages and they were able to find all the files I recorded, but track 1 and 2 seem to be 'mixed' together. Here is picture of the waveforms (every 2 lines is a pair of simultaneous tracks). The track with no mic hooked up recorded only hiss and it's now injected into the mic track at specific points.



Is it because the tracks were simultaneously written to the compact flash card that the recovery software can't differentiate between the data, or is this phenomenon a recorder+compact flash error (and unrelated to the format and recovery)? If so, I can rent the recorder again and try to recover from the internal hard drive. If not, is there anything that can be done?

Cheers,
William

the mix up file could be to do with the recovery software you where using
this happens a lot

i would try and get the unit back if the unit does not get written to again
then your audio file should be intact in the freespace of the hard drive

Thanks, I will try that!

The unit sets up 3 folders to organize the data when it is formatted... is that a big problem?

if the Audio file has what you recorded and the problem is only those additional tracks (channels)
i personally use GOLDWAVE, very Pro. Audio editing app.

you can remove those Hiss, and cut/paste and play around it to extract the recorded files

good luck

_________________
Kuwait Data Recovery - UNIX GTC
The only reason for time is so that everything doesn't happen at once. By: Albert Einstein

The hiss is so loud that it dominates the mic track in the combined segments, and since it is random, I believe it is impossible to eliminate.

I wish I knew the technical reason why these tracks got combined.
It happens at the same spots, so there is some reason for it. If only I knew more, like how to remap the clusters associated with the files.

this might happen with the App. you used.
why don`t you try another App. and see

Try R-Studio the demo ver. and compare the file size

_________________
Kuwait Data Recovery - UNIX GTC
The only reason for time is so that everything doesn't happen at once. By: Albert Einstein

I ended up using camera audio for this project (a lot of wind noise, it wasn't great) and so I put this on the back burner for a while.

I am still interested in recovering the mic boom audio.

I got the hard drive back before it was rented again but I had the same results as the CF card - the simultaneous tracks are merged at specific intervals, combining microphone boom audio and hiss from the unused channel into an unusable mess.

The main problem as I can understand it, is since the simultaneously recorded WAV files are the exact same length, they are being confused by the data recovery software.

Any ideas? Thanks.

Personally I doubt that the length of the files is causing that effect - more likely IMHO is the layout of the data on the disk, and the algorithm used by whatever recovery software you're trying. I've seen very different results using different recovery software on reformatted disks, for that reason.


Yes - but unless you become proficient in low-level filesystem structure examination & reconstruction, it may be unrealistic for you...

Anyway, If I was in your situation, I would make clones of both the original (reformatted) disk & CF card for safekeeping (not sure if you've done that yet), but I'd focus from then on with just one of them - probably the CF card as I'm guessing it's smaller.

Then I'd hire the same audio recorder again, set it up in exactly the same way as you had before (same folders etc.), make several recordings (as similar as you can do, to those which you had originally made) and make clones of both disk & CF card again - obviously before reformatting the unit ready for the next set of recordings. I'd suggest to make many sets of "simulated" repeats of what you did before, so that you can see if the audio recorder filesystem behaviour usually remains the same each time, by comparing these new recordings. Keep copies (clones) of them all, as you never know when they might be useful during the investigation.

Then look at the filesystem layout created by the recorder, on the media - especially the CF card, for comparison with what you are trying to recover. This will need experience of expected filesystem structures, and use of a hex editor etc.

That would be a starting point of iterative manual work, to investigate the layout of blocks on the "newly recorded" media, and which blocks get used, in which order, for which audio channel, and compare that with the "formatted" media you are trying to recover from. I doubt this would be quick, but I've had success in the distant past doing a similar process.

Another option for you is trying lots of different recovery software, as they may produce different results (as explained above).

Or see if you can get a DR pro to offer their services to recover usable files.

You asked for ideas - that's my $0.02. If that's useful then great; if not then please ignore

Vulcan,

Thanks for the reply.

Sounds daunting, but I may try what you suggested.

You wouldn't happen to know a good resource on fat file systems would you? Getting down to the nitty gritty?
I am engineer, so I might be able to tackle the subject matter, but no doubt it will take a while to learn.

Cheers,
William

Hi,

There are many sources for info on FAT - the "canonical" one from MS can be downloaded. This is the link for the document:

http://msdn.microsoft.com/en-us/windows ... 63080.aspx

Another place for info is this site (which explains FAT 16 & FAT32 - I guess your CF card is FAT32, if it's 4GB or larger):

http://home.teleport.com/~brainy/fat32.htm

There are many online sources of info, but it depends whether you want references, or tutorials etc. Have a search and see which style suits you, but in case of conflict between the docs, the MS document is probably correct of course.

Tools like WinHex (and some other hex editors) can automatically decode a sector as FAT and display the meaning of the various data fields, which obviously saves time.

I don't fully understand your description of files written by the recorder, but when you can duplicate the recordings, and see where each file is getting written (because the FAT will be intact), I hope things would become clearer e.g. does the recorder write 1 cluster containing data for one track, then 1 cluster for the other? Or some other, recognisible, pattern?

Another approach may be to try to identify the "hiss data" on the disk - if you can write your own software. I think that would be easier to do if the file format was uncompressed WAV, than if it was MP3, for example. I don't know which file format your recorder was using.

Good luck!