Fixable audio take at lower rate...?

[emfs2327]

Hello,

This is a general question I am hoping you can help. I recently have been experimenting with different frequencies and rates for audio recording using my STAudio cport2000 xlr jacks (two Apex 430 diaphragmatic mics recording into Cubase VST/32) and I did this recording of a choir in a church using 44.1kHz 16 bit at 512samples per second. The recording went off okay but with some problems that mirror the problem I have now when I try to add a bit of reverb. Here\'s the set-up: the choir does loud homophonic chords with silent breaks. In the silent breaks you can hear the choir\'s chord echo through the church. However, and I think this is because of the sample rate, you can hear in the echo after the loud chords a rat-tat-tat kind of effect, like the echo decays in ripples almost. That suggests to me that I should have recorded at a faster sample rate, and I have taken that note for this scenario with subsequent choir recordings. I now record at 96kHz with 2048 samples per second and then mix on down, and all my similar recordings since are free of this effect.

Now I try to mask the effect by adding reverb - specifically the reverb from Cubase VST\'s send effects menu (Reverb32, \'on stage\' if anyone cares.) I find that this reinforces the rat-tat-tat effect, not masks it, and so I am left with an artifact that could be quite useful but will not be unless I come up with a way to fix this audio. Can anyone suggest a technique or approach that will enable me to fix the audiotrack? I can\'t ask the choir to come back and re-record.

[Artspoke]

You seem to have an interesting problem here, and I\'m not sure that the sample rate simply being low is to blame - sound sort of like some screwy clock to me. I do not use Cubase, and therefore don\'t know what you mean by \"512 samples per second\". In the same line you said you were recording at 16 bit, 44.1 kHz - 44.1 kHz means 44,100 samples per second. Maybe if you can elaborate on the 512 part of the equation it will help.

Otherwise, to fix this problem, if I understand it properly, I would cut off that beautiful, natural churchrung reverb, where the artifacts seem to exist, put a quick fade to silence right after any portion of singing that\'s clean, and then put that track of edits through a nice warm reverb, trying to emulate the sound of the church. This way, you lose the actual churchverb, but with it you lose the problem spots as well.

I can\'t imagine that whatever is causing this anomaly during the \"silent\" part of the recordings isn\'t also affecting the singing parts. If you can\'t hear it there, then at least you have the meat of the recording - potatoes (the reverberation) can be simulated more easily.

On a hunch, check to see that your clock isn\'t set to sync to a source that does not exist - like word clock or spdif or optical or something. If you aren\'t slaving your system to any clock, it should be set to something like Internal.

-James

[Artspoke]

Oops, I forgot another thing I wanted to ask - can you see these false reverb iterations in the waveform representation of the track??

If you can see them, they were recorded that way, and the fix I mentioned above might be your only way out. If you can\'t see them, then it could just be a playback issue, in which case something else (still possibly the clock source) needs to be resolved, and your recordings might be just fine!

See what happens if you play these recording with a different program, or on a different computer. If they always sound the way you described, odds are the artifacts are in the recording itself.

-James