Vocals and compression

  • Thread starter Thread starter MBBoone
  • Start date Start date
M

MBBoone

New member
I just layed a few tracks last night. I was pretty pleased with the acoustic guitar. Given my Jimmy-rig setup, it sounded remarkable crisp and clear.

I am using the software: N-Track Studio. Very impressed with the product, especially for the price. There are a bunch of elements that can be tweaked that I am not familiar with.

For example the sampling frequency is set at 44100Hz - should I change this to 96000?

The only thing that I would like to improve are the vocals. They sound a little as if I am singing into an empty coffee can. Probably an empty can of Folgers coffee at that. I have seen on other posts, that by manipulating the compression, that it could possibly improve the clarity of the vocals - is this true? If so, what would be the recommended settings a 1:10 ratio or 10:1 - I don't know the difference.

thanks,

mb
 
I think you'll be wasting some HD space by selecting 96 KHz instead of 44.1. Consider this: Most humans can't hear squat above 16KHz. A few can hear up to 20KHz. Now making the rather crude assumption that your sample rate should be twice the highest frequency sound you intend to capture leads to the conclusion that sample rates above 40KHz are a waste of resources. 44.1KHZ is really convenient, because standard audio CDs are burned at this sample rate. Extending the sample size from 16 bits to 24 or 32 would be the first thing that you'd want to do if you've got more resources than you can use. Here again you still have to deal with the conversion to 16 bit for the burn to standard CD format.
As to compression: I'm really new at this as well, but I'd play with ratios in the 1.5:1 to 2.5:1 range and also adjust the threshold
value in accordance with the signal level on the track you're compressing. The effects are very subtle, so have a really good monitoring system, take your time and listen carefully.
 
Yikes!!! What a barrel of monkeys you have opened here. As I always like to say, "If you open a barrel of monkeys, be prepared for a gorilla to jump out at ya". :) But since we are talking digital audio here, my recent trip to the zoo really doesn't have much pertinence now does it? Or maybe....

Anyway. drstrawl has given you half the information concerning what we can hear, and what we can't. Actually drstrawl, the highest frequency a human was ever tested hearing was about 18Khz. Next, just because you can't hear it doesn't mean that it doesn't have a major effect on the sound. Trumpets create frequencies up around 28-30Khz range. You may be thinking to yourself that that is almost a full octave (double the frequency) higher than you can hear, so who cares? Well, physics care!

You see, it is the upper harmonics, or the overtones of an instrument that give that instrument it TIMBRE, and I don't mean lumber or cute trees here, I am talking the actually perceived sound of the instrument. Have you ever asked yourself why two instument that are seemingly exactly the same, playing the same note sound quite a bit different. You can blame it on the player, but there are things called Phycoacoustics in play here. Now I am not suggesting that sound is crazy (well, on second thought......)phycoacoustics have to do with perceived sound, and what has an affect on it.

On we go here. The reason we concern ourself with upper freqs is because they give instruments their very distinct sound. If you can't capture those freqs, then the accuracy of the sound is diminished. In the case that you can't hear those freqs, it doesn't matter because it more has to do with the way that those freqs mess up what you can hear. So, we need to have these freqs present to mess up the sound for us so that everything sounds right. Almost doesn't make sense to have something messing everything up and calling it the right thing to do, but that is the case here. Without those little weak upper freqs clashing around, the perceived sound is less detailed.

Now things get tricky. You may be asking yourself why most audio components only go up to 20Khz if you need the stuff much higher than that. Of course you could also ask the logical question of why a recorder would need the ability to record higher than your ears can hear. And these are very good questions. But, if you think like a microphone and a speaker, you will get much better result in the studio.

Here is the deal. If somehow you could just put yourself inside of your digital recorder and hear what is there, man, that would be some sound. Unfortunately, we can't do that so we rely upon speakers to do the exact opposit of what a microphone does. In neither case do they do this as well as the ear. The ear is one sensitive microphone. And the brain is the ultimate speaker. This must sound crazy, but really, I am not high on anything, just using some color for some very dry stuff here.

So, since a microphone has to pick this stuff up, and a speaker has to reproduce it, and since neither is as good as the ear, we need to overkill a few things here to help out. Actually, most audio equipment does have a frequency response that is quite large. Manufactures only concern themselves with telling you how well they respond to frequencies that you CAN hear. But when you start reading up on the phase shifting of the stuff you can hear, that is where you can make pretty intelligent choices about the equipment you purchase.

So let use the 44.1Khz sampling rate here. It is a fact that a you can only record a frequency half the sampling rate. Soooooo, if the music contains information that is much higher than this, well, you are missing out on the stuff that gives it realism. It is as simple as that. The reason that sampling rates on professional recorders is standard at 48Khz is because is sounds better. More of the detail in the sound is picked up and reproduced accurately. So in the case of 96Khz, we would definately be able to record ANYTHING that might affect the perceived sound. That is why things are moving this way. So, if you have the hard drive space to record at this sampling rate, by god, go for it. You definately will hear the difference. As far as the material getting downward sampled to 44.1, well, you are still going to wind up with a better sound on the CD.

Next. Bits o' resolution really have to do with variances in the volume of the frequencies. More bits gives more procession power to dedicate to reproducting tiny fluctuations of volume that were picked up by the converters. More information can be stored on tape. So, go with the highest you have if you have the hard drive space.

Maybe play around with say a 88.2Khz, 20 bit recording for the source tracks. This will give you much better fidelity than is currently available on many widely used recording formats. That is, if it is true linear 20 bit. Watch out here, many products don't deliver true 20, or 24 bits of resolution. They use some data compression schemes to simulate it.

Next. A compressor on vocals can actually lead to high lighting an already bad sound by making it more present in the mix. If your vocals sound like they are coming from a Folgers can, well, first I would ask why with all the Starbucks in the world, you are still drinking that crap.... :) Next I would suggest that maybe you are not using a very good microphone or microphone pre-amp, or that possibly you have some cable somewhere in the signal path that is bad. Compression will not alter the tone of anything severly unless used in excess. So, try borrowing a really good large diaphram condenser mic, and a decent outboard mic pre-amp for your vocals, and then see if that will improve things. Also, when mixing a vocal, about the only thing you will need to do with it on the EQ is to roll off some low frequency if a high pass filter wasn't employed while you recorded the track. MAKE SURE THAT YOU ARE GETTING THE RIGHT KIND OF TONE AND PRESENCE ON THE VOCAL WHILE YOU ARE RECORDING!!! Any EQ'ing on a vocal is usually to it's detriment.

Hope this helps. Good luck.

Ed Rei
Echo Star Studio www.echostarstudio.com
 
>Yikes!!! What a barrel of monkeys you have >opened here.

No Shit! And Ed, I'm not questioning your considerable ability behind the recording desk. Just a quibble over digital resolution levels with which I have no direct experience. Nothing but academic exposure in acoustic physics.

>Actually drstrawl, the highest frequency a >human was ever tested hearing was about >18Khz.

20KHz was an optimistic estimate. And I'll bet that dude never played live while still retaining that 18KHz frequency response.

>Next, just because you can't hear it doesn't
>mean that it doesn't have a major effect on >the sound. so who cares? Well, physics care!

If you exclude such frequencies up front by limiting the ability of your digital system to record them, they won't come into play as the bandwidth hogs that you purport them to be.

>phycoacoustics have to do with perceived >sound, and what has an affect on it.

I'm just not convinced that waveforms that are ignored by my speakers will significantly come into play in the perceived sound of my mix.

If I had resources to spare, then I'd expand my system in both directions; namely sample rate and sample size.
I was just saying that sample size will have a more noticeable effect because that allows the quieter parts to be that much quieter and creates more distance from the bottom to the top of the dynamic range, which is audible over the entire audible frequency range.

>because it more has to do with the way that >those freqs mess up what you can hear. So, >we need to have these freqs present to mess >up the sound for us so that everything >sounds right.

If these frequencies are excluded up front, I think you still have enough high end to get a crisp mix.

>And the brain is the ultimate speaker. This >must sound crazy, but really, I am not high >on anything, just using some color for some >very dry stuff here.

Works for me!

>Actually, most audio equipment does have a >frequency response that is quite large. >Manufactures only concern themselves with >telling you how well they respond to >frequencies that you CAN hear.

I'd suspect them of hooey if they felt otherwise. At least in the role of the near-field monitor. Consumer speakers are charged with the task of fixing all the ills of a cheapo front end. Monitors are charged with "telling it like it is."

>So let use the 44.1Khz sampling rate here. >It is a fact that a you can only record a >frequency half the sampling rate.

Well at least we agree on that.... :)

>So in the case of 96Khz, we would definately >be able to record ANYTHING that might >affect the perceived sound.

The idea of both 24 bit and 96 KHz is a come-on to consumers that can't afford real methods to improve their recordings. Marketers have predicted (and I'd agree) that hardware will achieve this quality and price reduction in a consumer system way before the price of a good condenser mic changes at all. So to be on the bandwagon they push the higher resolution stuff without a real reason for the higher resolution other than that it can be achieved on the cheap. Plenty of headroom; filled with too much noise!

>That is why things are moving this way.

As per above, I disagree.

>So, if you have the hard drive space to >record at this sampling rate, by god, go for >it.

I DEFINITELY agree. You bought it. Why not use it to the maximum extent possible? You can always listen up, and later agree with me and increase the # of minutes of 16 tracks you can play/record without losing anything.... :)

>You definately will hear the difference.
>As far as the material getting downward >sampled to 44.1, well, you are still going >to wind up with a better sound on the CD.

Please let us know about this.... :)

>Folgers can, well, first I would ask why >with all the Starbucks in world, you are >still drinking that crap....

The only real Finca cafe is from Carl Diedrich. I'll dig up the web address if you want. He's got his own plantation in Guatemala. He began by bringing back 50Kg. bags of green beans to Costa Mesa in his VW van. He recently turned down a $17M buyout deal from Starbucks! I've known him since 1979. His beans ROCK!
 
Sonusman took the words right out of my mouth. Or the essay out of my head. Whichever.
 
Hey...uh.... hook me up with that coffee guy address. I'm going------------>

[This message has been edited by Lorddiagram (edited 09-30-1999).]
 
Back
Top