Skip to main content

(Licensed by CC Protection BY-NC-ND)

http://www.megaupload.com/?d=UQKNB1FT

It sounds a little bit like I'm hearing someone sign karaoke, not because of the quality of the voice (i think shes amazing) but because of the quality of the recording... am I EQ'ing wrong? I hear it especially in the beginning it sounds way to "live" and not like another instrument in the song. why?

And oh, yeah.. how do you like the song? Comments?

Topic Tags

Comments

anonymous Sun, 09/13/2009 - 09:33

OK, I've listened under headphones and also through nearfield monitors.

I agree with Guitarfreak that de-essing would probably help. The high frequencies are hot and producing lots of MP3 artifact, which you wouldn't hear on the master, of course. Still, based on the style of music and target audience, I'm guessing that MP3 is going to be the final product format, so cleaning up sibilants and transients before they wreak havoc with the compression encoding is going to be critically important. It's unfortunate, but that's the world we live in.

Overall, I would not decribe the vocal as "live" at all, in the reverberant sense. If anything it is too dry. The guitars and drums are spread fairly well across the stereo space, but the vocal is completely mono, panned to various points. You are moving the point around, from verse to chorus, and backing to refrain, but it's always a single point and lacks both width and depth. What to do about it is probably a matter of taste.

Personally, I would not add conventional reverb, as that would sould too old-fashioned for this style of music. I'm thinking that adding in some delayed copies of the vocal, probably with more highs filtered out, panned left and right (not all the way) would provide both some width and depth. I would probably soft-gate them, as well, so that the delayed signals don't jump out at the beginning and ending of each phrase (open the gate slowly and close it quickly).

The quality that might be described as "live sounding" (which you may be associating with a "karaoke" sound) seems to be a resonant peak very close to the bottom of the vocal range and a sharp hi-pass rolloff below that. I don't think it's the singer--it could be the microphone or something in your EQ. Perhaps you were too aggressive with the lo-cut to kill pops. One reviewer mentioned hearing the room; I think I do detect a bit of that occasionally, but I don't think that's the central problem.

Assuming that you used a good-quality, fairly neutral vocal microphone, I would have to ask, what _are_ you doing with the EQ? It seems that all the vocals have some sort of filter effect; obviously, some parts are filtered more than others, but they all seem to have that resonant peak. Is that your intent? By my way of thinking, if you're going to use a filter for a "telephone" effect, you either want it in or out, rather than keeping it in throughout the song and varying the Q of the filter between verse to chorus (and adding distortion for the refrain).

There's a couple of spots where the levels are ducked a little too much (instruments at the beginning of one of the verses and the vocal on one of the refrains panned right, if I recall correctly), but that's not a big deal; mastering should even that out.
--
David
Sugar Land

mannyr Sun, 09/13/2009 - 12:25

I'm using a Rode NT2 microphone.

What would you propose I do with the EQ so that this resonant peak could be taken care of? Where in the frequency range is this resonant peak you hear? And what do you mean by hi-pass rolloff?

I want the vocals to sound like they have more presence, and that they sound like there very neutral, in your face and one with the song.

The resonant peak wasn't intended. Do you really want that delay that you mentioned to happen throughout the whole song or just the chorus? Any other ideas how I can add more depth to the vocals?

What do you mean I was to agressive with the lo-cut? I cut too much lower frequencies out? What do I do with this EQ?

I'll

anonymous Sun, 09/13/2009 - 15:43

What would you propose I do with the EQ so that this resonant peak could be taken care of? Where in the frequency range is this resonant peak you hear?

Somewhere around 500-1000Hz.

And what do you mean by hi-pass rolloff?

High-pass is the same as "low-cut", so what I was referring to was low-frequency rolloff. (I'm a EE, so filters are all "low-pass", "band-pass" or "high-pass" to me). Your mic has a built-in high-pass filter that is switch selectable for a cutoff frequency of either 50Hz or 80Hz. Since this is a vocal, neither position should have any effect on this material, other than suppressing really low pops and breath noise. I was wondering if you were cutting any frequencies higher than that, in the range up to 500Hz. The lowest note of the vocal on this track is about 250Hz; so we don't care below that, but you don't want to lose any of the fundamental frequencies in the voice.

The NT2 has a pretty flat response. Compared to dynamic and ribbon vocal mics, it does not have much of a presence peak between 2kHz and 8kHz (practically none at all in omni mode). Like many condensers, it is quite bright above 8 kHz and that needs to be tamed in many situations.

Rather than a resonance, it may be that what I'm actually hearing is the combination of a lack of proximity effect in the octave from 250-500Hz and a lack of a presence peak in the octaves from 2 kHz to 8 kHz. That makes the range between 500Hz and 2kHz stand out, giving a "small box" sound.

If you had the NT2 set on the omni pattern, there would be no proximity effect at all. Otherwise, how close was the singer to the mic? You might want to boost the octave centered around 250Hz to compensate for the lack of proximity effect.

Some presence could be added with a broad boost in the octaves from 2 kHz to 8 kHz. A parametric filter set to a fairly low Q might be best for this, as you can try different center frequencies to find what works best for this particular voice.

Finally, at the high end (over 8kHz), you could use a de-esser or some other fast high-band limiter to tame the sibilants. Simply taking the high end down a few dB might help take the edge off, as well.

Naturally, it's best not to have to do all this fiddling with filters, but sometimes you have to play with the hand you're dealt.

Do you really want that delay that you mentioned to happen throughout the whole song or just the chorus?

I would use a delay-based reverb throughout, but just a touch. Too much will begin to make the vocal more distant, which is not what you want. Having the presence boost on the direct vocal and not the reverb will help keep the voice up front. You definitely don't want the sibilants ringing in the reverb, as that will call attention to it. I do think the added depth will help achieve better blending with the instruments.

Using it only on the chorus would be more like an effect, and the chorus has plenty of effects going on already.
--
David
Sugar Land

hueseph Mon, 09/14/2009 - 00:46

First things first. Turn it down. The whole file is compressed to oblivion. There are no dynamics. Real music doesn't sound like that. Not to say that your song is not real music. The song has potential. It needs space. There needs to be some air between the singer and the listener.

I have my monitors down quite low and it still sounds like you're in my face. There is so much distortion on that track that it's hard to make a clear judgment. I honestly don't think the vox sound that bad but, it's hard to say. I think if you could tame the rest of the mix, the vox could cut through the mix quite nicely.

There is a general deficiency around 100hz which is unusual. That doesn't mean you should go adding 100hz to everything. Maybe add a touch to the bass to fill that out. And, give the other instruments a space of their own. Would you be opposed to putting the individual tracks up and letting someone else mix it for you?

At any rate, I think the song is great. Lot's of potential. Just a little overprocessed.

anonymous Wed, 12/02/2009 - 22:31

the effects parts should be lower in the mix, like WAY lower (- 5db, if not more). her voice is great. it is very present though, almost too present (present meaning pushed out in the mix).

reverb and a wide Q mid cut (not too much, maybe a 1-3db cut) would let it sit a little better in the mix for my tastes. a little trick i like to use for a smoother sounding reverb is to stack reverbs.

lastly, it sounds like you may be getting some clipping. recorded too hot perhaps??

x

User login