Karaoke Forums

I've gotten a lot of comments in the Singers Showcase on the quality of my subs so I thought I would take a few minutes to lay down how I do it. First off, although I have both Windows machines and a Mac I do all my audio work on the Mac. Why? Apple tends to cater to the pro audio and pro video markets so a lot of the stuff that is out there is extremely high quality, even software that you would think would be cheap or is included with a new Mac purchase.

The foundation of any good sub has to be what you are using to capture your audio and that boils down into two different areas, the music track and the voice track.

In order to do the music track in THE best way you need to rip the audio from the CD you are using (sorry, tough to get a quality sub using a tape backing track). I suggest ripping to the native, uncompressed format -- for Windows this is WAV, for Mac it is AIFF (although it will also do WAV I recommend AIFF). I won't go into rippers, there are a lot out there. From there you need to add the music track directly onto a track in your mixing program (discussed later).

For vocals you're going to need a mic and an input device of some kind (commonly called "audio interface hardware"). There are a number of interfaces available on the market, personally I have an M-Audio Firewire 410 which gives me two XLR inputs with decent preamps as well as phantom power (you can find these for under $200 out there now, believe the model is being phased out). USB devices are also out there, the Lexicon Omega is a great unit with a decent mixer packaged in and there are some other quality and inexpensive units from TEAC and TASCAM. M-Audio, Presonus and others occupy the midrange with Digidesign being the Mercedes of audio interfaces. Preamps are very important as they are the #1 cause of noise and poor performance with microphones.

I suggest Firewire because drivers tend to be a lot more stable and you get a lot more bandwidth with Firewire which means Firewire tends to have more channels available (hence my M-Audios 4 inputs and 10 outputs). Firewire tends to be a bit more expensive on the whole.

As far as mics there are a number of options. You can get a simple dynamic mic -- Shure, EV, and AKG all make decent dynamic mics that are not all that expensive. Again, if you want the best sound though you'll want a condesner mic (hope that interface you got has phantom power, must have to use condenser mics). Among condensers I have been told the Studio Projects C1 is a cheap mic ($200) that holds its own with those costing 5 times the price. You'll need a decent XLR microphone (balanced) cable to hook it up to your interface.

Okay, time to hook it all up. You connect your mic to the mic cable which connects to your interface. The interface connects to the computer ONE WAY, and that is via the Firewire or USB cable depending on the type of box you got. You DO NOT NEED to hook up audio outputs from your interface to your soundcard (#1 mistake I see). Think of your audio interface REPLACING your sound card (that's exactly what it's doing).

Now, mixing. Remember what I said about Apple? Well, on every new Apple out there, including the recently released $499 Mac mini, they include GarageBand. GarageBand is an outstanding mixer that is easy to use and has quality effects with tons of can't miss presets all included. If you're in a Windows world Cubase is the most common (and included with some of the interfaces so consider that when shopping for one).

Let's say you're using GarageBand, you open it up and create your first track which is a "live instrument" and select "basic track" and you'll see a selection called "no effects" and select that. Now, open up the Finder and drag your audio file you made for the music track and drag it to that track you just created. Voila, audio track added directly to your project, no conversions, no cables to go through, pure and clean digital audio.

Now, we're going to record your vocals. First, check the preferences in your mixing program and make sure that your audio input is set to your audio interface device (in GarageBand select "preferences" from the Apple menu, then "audio/midi" and then make sure you have your audio interface selected for the input.

Next add another track to your project, in GarageBand select "real instrument" and then "vocals" and pick a preset that suits you. Later on you can play with this some more but "live performance" is a good general use setting with a light amount of reverb and some compression. Make sure you have the correct input for your mic and that it it set to "mono" (most of the time this will be "Channel 1 Mono" in the drop down box) that will put your voice in both left and right channels. Finally, you can select your monitor setting, I suggest hooking up your headphones and turning it to "on" so you can hear yourself.

Now, how are we going to record this baby? There are two ways. The first is to have the lyrics available to you and record the vocals using your mixing program playing the music track. This way works well if you know the tune cold.

If you don't know it cold then you're going to need to open up your karaoke player to do it. If you have a computer karaoke player (most common on windows is Winamp with CDG plugin, Mac is kTunes) this is best if you ripped your track from CDG. Even if your karaoke player is hooked up elsewhere that will work as long as you're singing into the computer's mic and recording from there, makes no difference how you playback when recording because we're only recording your voice. This method works but you might have to go back and sync the vocals with the music which can be a pain. So, I suggest the first method if possible.

Now, you've recorded your track, in fact, you've gone back and redone sections if you needed to, and you're going to touch it up. Make sure you remove the opening and ending silence where needed (a simple drag on the beginning of the track in GarageBand) and in pauses in your vocals. This will eliminate the noise from when you cleared your throat or your kid came and asked you what you were doing!

Check your vocal levels and make sure they're not too hot (loud). Play with your reverb some until you have it set just as you would like (less is usually more), play with some presets to make it sound better. Finally, add some EQ to the music track to make sure it is nice and lively (in GarageBand I recommend the "hi-fi" EQ setting for the music track).

Now save your project so you can go back and play with it if need be and then export it. In GarageBand you export to iTunes and you'll see iTunes open up with your new track. From there you can convert it to MP3 (go into "preferences" in iTunes and set to "MP3" for importing and change your MP3 settings). I recommend 128 VBR, high quality for most songs under 5 minutes, this can be different depending on how big you want the file to be and how long the song is. 160 is the optimum MP3 setting without many artifacts.

And, there you have it, quality subs without much noise, almost no hum or other artifacts and it didn't cost you an arm and a leg. I'll be happy to elaborate more if you want some more nitty gritty.

thanks for the info