FM voc, sharp 'whitsle componant?!

  • Thread starter Thread starter mixsit
  • Start date Start date
M

mixsit

Well-known member
Ever run into this? Working on this lady's vocal, in a few dif songs I've noticed there isn't a lot of highs or presence to begin with. But in some songs, or places in songs, what presence tone there are there is mostly a rather narrow band of highs'. In one case I had dialed out a 4.5k notch. Later, another song, there was a 'whistle component a little below 4k - in a spot with a strong vowel sound no less- repeated pretty much every time in that phrase ('word actually.
I've had vocals that have nasty edges' and stuff, but this is weird. A 'covered tone, but with this in it?
 
That was a bit strange....I don't understand either. Maybe a few cocktails for it to make sense. :)
 
Gads, Ok. Starting again. Begin with a voice that has sort of a covered' tone- i.e, not much of upper mids to it.
Now if you sweep around looking to find what's available in the way of presence tones that can be brought out, in this case what I find is predominately (in the first case) is a very narrow narrow band up around 4.5k. From another perspective- listening to it flat and taking that as the baseline', there would be this feather edge poking out at 4.5k several db. This was on a few songs.
On the other example- this happening, a little lower- sticking out on a vowel (not on an ess' for example.
Mic was an re-20 FWIW.
 
You need a tube mic and a tube mic cable and Apogee interface to get the most out of Audacity.

And more room treatment.

Try making her sing with one of those anti-tooth grinding rubber things in her mouth. Should help with sibilant bowels.

Do some takes at 24-bit/196k resolution and post the raw WAVs so we can hear what you're talking about.

Wait, it's the phantom power on your USB mic crapping out again.

Try compressing the USB mic. Use the USB insert on the USB mic.

:laughings:
 
Strange, but the effect must be coming from her or the room, unless the mic has developed a problem. Did you try recording her with a different mic to see if the problem repeats? Maybe not the best mic choice for vocal singing (assuming this is not voice-over work).
 
Sounds like a resonance of some sort. Could be that the prominent aspects of the vocalist are exciting the weaknesses in the response of the mic. I doubt resonances up that high are a result of the room.

If you're recording her, try putting up some more dampening and try a different mic.

If you're just mixing, use a dynamic EQ (the Platinum Ears IQ4 VST (windows) is free) to squash those resonances. I have quit notching altogether as it's too static. A dynamic EQ is a far better way to fix these problems.

Cheers :)
 
If you're just mixing, use a dynamic EQ (the Platinum Ears IQ4 VST (windows) is free) to squash those resonances. I have quit notching altogether as it's too static. A dynamic EQ is a far better way to fix these problems.

Cheers :)

Whoa. Thanks Mo. Just tried this on a 'clanky' bass track I was fighting with. Problem now solved. :)
 
> I love IQ4 the way it handles selective tone shaping- though I kind'a fell away from it being a bit glitchy in Sonar. This gal I'm using de-whistle' notch automations. :D
I'll get a chance in a few week here to do some more tracking and stuff. Now that I'll be watching for it it should be reveling.
 
So the problem lies at the source?
I think so. Same mics-(a pair of re-20s in play, but between the original band tracking and some overdubs I haven't been keeping track of which one was 'voc and which was 'bass.
One interesting bit, maybe-- My wife made a comment that her voice -from a fm perspective?
evoked a baby cry' response (i.e. 'not pleasant? IDK, I thought it was interesting but not related to this particular thing probably.
 
The RE20 has virtually no proximity effect so this might be the cause. It may just not be right for her voice.

In any case, maybe try a darker mic.

And @Jimmy, no worries. Glad it helped you. It's a great processor and I use it often for fixes such as this. The thanks goes to Dan Worral for coding it and making it available for free. Is he on here? I think he might be on recording.org. You can really dig into those ringing frequencies without the static, "cloth over the speaker" effect of heavy notching in the upper registers. It's good for just about any problem of single band-related dynamic swing.

Cheers :)
 
The RE20 has virtually no proximity effect so this might be the cause. It may just not be right for her voice.

In any case, maybe try a darker mic.

And @Jimmy, no worries. Glad it helped you. It's a great processor and I use it often for fixes such as this. The thanks goes to Dan Worral for coding it and making it available for free. Is he on here? I think he might be on recording.org. You can really dig into those ringing frequencies without the static, "cloth over the speaker" effect of heavy notching in the upper registers. It's good for just about any problem of single band-related dynamic swing.

Cheers :)


I actually forgot about that forum. I will thank Dan personally. :)
 
Sigh...another similar lame search engine with this forum software. Funny how you can do an advanced search for a member, and it prompts the user name-yet no results for the search. Seriously lame...
 
Get a linear mic, never mind all those "vintage" mic types, that is just saturation and EQ, one can really add later. And a good multibit D/A. Singlebit sometimes sound like 4bit, because it does not have the neccesary bandwith to accurately reproduce the signal, but it does filter out most audible distortion. That does not equal 24bit though ;)

And if there still should be problems, and this trick you can use anyway, use a singleband limiter, set at a fast release, and eq after, to remove the unwanted frequencies, and try to make up for the loss of wanted frequencies with an eq before. This is also a good de-esser.

Peace Be With You.
 
Get a linear mic, never mind all those "vintage" mic types, that is just saturation and EQ, one can really add later. And a good multibit D/A. Singlebit sometimes sound like 4bit, because it does not have the neccesary bandwith to accurately reproduce the signal, but it does filter out most audible distortion. That does not equal 24bit though ;)
RE20...
And if there still should be problems, and this trick you can use anyway, use a singleband limiter, set at a fast release, and eq after, to remove the unwanted frequencies, and try to make up for the loss of wanted frequencies with an eq before. This is also a good de-esser.

Peace Be With You.
As tempting as it is to jump to a multiband level detection' fix for something like this there's a slight flaw in that application.
Compare it to 'de-essing for example. There's a little excess energy up and you want to bring it down some so it's back in balance'. This is not quite that. This is a sharp 'whistle that doesn't belong in the middle of a vowel. This solution needed the gain reduction to go something like -9 below the surrounding level.
See a difference there? :)
 
Back
Top