Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Unfixable Click/Pop? (Read 1547 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Unfixable Click/Pop?

I didn't see a specific subforum for a topiclike this, so if I overlooked it, please let me know.

I'm trying to remove a click/pop from an audio file of mine, and it's not working. When I fix it manually with Audacity, it just moves, and when I try to fix it with ClickRepair, it only makes it smaller, but still quite noticeable. Since the song loops, the noise is present a few times throughout the file. Is there anything I can do?

Re: Unfixable Click/Pop?

Reply #1
Try deleting a few milliseconds around the click.

Re: Unfixable Click/Pop?

Reply #2
If only in one channel, replace it with what's in the other channel.  If in both, replace it with something nearby.

Re: Unfixable Click/Pop?

Reply #3
Firstly, I forgot to post the picture I made to go along with the post initially. Here it is.

Try deleting a few milliseconds around the click.

This happens. Essentially, the pop just moves.

If only in one channel, replace it with what's in the other channel.  If in both, replace it with something nearby.

The click remanins. It seems to adapt to whatever I do to it.

Re: Unfixable Click/Pop?

Reply #4
Try using the  the Draw Tool in Audacity to manually re-draw the waveform.

You also try Wave Repair which offers several different repair algorithms.   It work best when you manually find & mark the defects.  (There is a 30-day free trial.)


..But, You can't always fix everything.

 

Re: Unfixable Click/Pop?

Reply #5
That defect looks like a skip, where a section of the signal has been cut out, perhaps a dropout during recording, rather a click perturbing an intact waveform. You might have to delete a long section surrounding the click to avoid a low frequency thud/pop in either channel. Try selections of different length such that the waveform lines up at the selection's start and end ( such as if both are near 0 ), listen, undo and try another one. You might want to try a crossfade between the portions before and after the click. Again the crossfade needs to be quit long in order not to be perceived as a thud.

iZotope RX Declicker can sometimes successfully conceal clicks that seem unrepairable. If you uploaded a sound file, I could give a try fixing it with Declicker.

Re: Unfixable Click/Pop?

Reply #6
Yes, it certainly looks like a lousy edit.   If you have editing software that lets you do crossfades at the edit point, set the edit point at the discontinuity step, and try a fairly long crossfade, let us say 100 milliseconds or so.  If you have a spectrum editor, you can suck out any low frequency content issues around the edit point    Yes, Izotope RX5, particularly the pricey advanced version can probably clean this up well, in any of several ways, although if you want to change the crossfade time at edit points you have to set it in the options settings for each edit.

The click you hear is often referred to as a Fourier click,  The step at the edit point essentially is a momentary burst of high frequency content in the middle of what is otherwise low frequency material.  Google Fourier Transform.  Good editors have settings to only edit at zero crossings to usually avoid this. 

One other technique which would probably work is to momentarily roll off the low frequencies either side of the discontinuity Not sure what would be least audible.

In the days of tape and razor blades, avoiding Fourier clicks is why splicing blocks made the splice at an angle.

Re: Unfixable Click/Pop?

Reply #7
It looks to me like if you simply delete both the samples from your highlighted section and all samples back to about 4:01.873 (or wherever the amplitudes match the best) you would get a continuous waveform in both channels. That would be much easier than trying to duplicate what was cut out.

Re: Unfixable Click/Pop?

Reply #8
Firstly, I forgot to post the picture I made to go along with the post initially. Here it is.
The audible effect of clicks can often extend quite far beyond their visible appearance in a waveform. I suggest you select a dozen or so samples before and after the click and interpolate. Given the simple nature of the waveform, even a simple linear interpolation will probably work.