0

I have a cassette tape recording of a spoken radio interview. The first half of the recording went bad years ago. Now, whenever I play it (regardless of the tape player), it sounds something like this:

https://soundcloud.com/josh-renaud-1/interview-damaged-section/s-fXRT28ADNZ6

The damaged portion is 14 minutes long. As the recording plays, it sort of recovers gradually. Finally, at a station break, the audio finally becomes perfect again. I have about 6 minutes of good audio after the damaged 14. The audio clip I shared above is from the end of the damaged section -- the beginning sounds much worse.

Can anyone tell me precisely what is wrong? Is it "distortion"? Did the tape get stretched or something?

And is there any way to fix the audio digitally so that the voices at least become understandable, even if the quality remains poor? If so, what filters or techniques should I use?

(I own Amadeus Pro for the Mac, but don't know much about very technical audio work)

Improve this question
3
  • 1
    It's called wow… though in this particular case it may be WOW ! ! ! ;)
    – Tetsujin
    Mar 19, 2020 at 9:05
  • 1
    Actually, I just listened to the 2nd example. "Wow & flutter" usually come as a pair - that 2nd one is the epitome of flutter, gone mad. I really see little hope of fixing that.
    – Tetsujin
    Mar 19, 2020 at 13:31
  • "Wow and flutter". I never heard of these before. Thanks so much for helping me, I learned something new.
    – Kirkman14
    Mar 19, 2020 at 21:33

1 Answer 1

Reset to default
1

This is a problem that occurred during the original recording. Sounds very much like the tape capstan rubbers are slipping and the tape is not being kept up-to-speed during the recording.

There is only one piece of software that has any hope of fixing this, and it's "Capstan" from Celemony (makers of Melodyne). Make sure you are sitting on a good solid chair before you search for this on their website.

Improve this answer
10
  • 1
    Sitting down, with a good strong drink in your hand ;) Capstan might be able to do it, but by heck you're going to need some patience. It will be a manual curve-drawing exercise. Capstan really likes complex music for best analysis - the less info it has to work from, the worse job it makes. A solo speaking voice is about as bad as it gets.
    – Tetsujin
    Mar 19, 2020 at 8:55
  • 1
    ...and the clip the OP posted was the good part... :-(
    – Mark
    Mar 19, 2020 at 11:43
  • 1
    I just gave the first example to Capstan - it thought about it for quite a while & eventually came up with… no improvement whatsoever :\ It has nothing to track, really. [What can I say, every single movie & TV project I had lined up for the next 3 months has been cancelled or postponed… I've not a lot else to do except mess with stuff to see what happens ;)
    – Tetsujin
    Mar 19, 2020 at 13:33
  • yeah we're all in the same boat mate. Might be worth posting a Stack question in this light and see what people come up with!
    – Mark
    Mar 19, 2020 at 14:34
  • We could start up the the chat room here… "Hey, whatcher doin' today?" 'Pickin' ma toenails & watching daytime TV." "Yup, me too…" What should you have been doing? I had the next 3 months booked on the new H.Potter, Fab Beasts 3… They even cancelled Eastenders, which is my usual bread & butter :\ [& I just this minute found I will get paid for today's cancellation, so Yay!
    – Tetsujin
    Mar 19, 2020 at 15:33

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.