To the Blog
Jason Barker is Currently...
Creating GET WISDOM 3: Choosing a Channel Strip
Posted on Monday, November 12, 2007 at 3:53 PM by Jason Barker
In the first two articles in this miniseries, I've written about the criteria and processes I used in selecting a microphone and an interface to create GET WISDOM. In this article I want to briefly discuss the channel strip I use, as well as reasons for not using a hardware audio processor.
Before continuing, I want to emphasize - as I have in my other articles - that the choices I've made are based upon my personal preferences, and are not indicative of any official recommendation from the Department of Youth Ministry or the Orthodox Christian Network. You may find that different choices in equipment or processes may better suit your preferences or circumstances.
A channel strip - also commonly called a mic processor - provides audio processing. This means that it processes - in other words, alters (and, according to taste, improves) - the audio that is recorded. A channel strip generally includes basic versions of several audio effects that can be performed by more sophisticated and powerful - but also more expensive - individual units, such as compression, de-essing, gating, etc. (I'll briefly discuss these effects below).
There are two general approaches to processing the audio for podcasts. One approach - and arguably the one most often recommended for beginners - is to use software for your processing. This is a very valid approach, and one that is relatively easily accomplished because most audio recording software offers at least a small array of basic processing effects, and there are also a number of free or inexpensive individual options (such as the very popular Levelator). The advantage of using software for your processing is that our original recording is pristine, i.e., unchanged by any effects. You can therefore try and retry different effects and settings in your postproduction (the editing you perform on your audio after recording it) until you have the audio track sounding precisely the way you want (or, in my case, sounding as good as you can make it until you finally get sick of postproduction and decide to just finish the episode). Also, if you make some "unfixable" mistake in postproduction, you will still have the original audio track upon which to fall back. Another important factor is that, since basic processing effects are built into most audio recording software, you will not need to spend additional money on a hardware processor.
The other approach to processing is to use an external hardware processor or processors. Many fully-dedicated professionals will use individual compressors, limiters, etc., to achieve the greatest control over their audio, but most part-time professionals or amateurs (like myself) who use external processing will use a channel strip that combines basic versions of this equipment into a single unit. The advantage of using a hardware processor is that you can hear exactly how the recording will sound as you're recording it (and even before you start recording). This greatly reduces the amount of time you'll need to spend in postproduction, but it also means that, if you have some problem with your audio that you didn't notice when you were recording it (such as a background noise that you didn't gate, or accidentally overcompressing your audio), you will either need to re-record the audio track or be stuck with the problem in your audio file.
Ultimately, your decision as to whether to use software or hardware audio processing will be based upon what is most comfortable for you. I am more comfortable using a hardware processor, because I prefer being able to monitor how my audio sounds as I'm recording. I - like many people - experience significant problems with latency when recording, meaning that time it takes for my computer to process and record the audio means that what I hear coming from my computer when recording is slightly behind my speech; by plugging my headphones into my interface, I can hear precisely how the audio sounds before it reaches the computer. Also, because I can leave my channel strip at the same settings for my recordings (with slight adjustments for current recording circumstances), I save time on postproduction.
One of the most popular channel strips for podcasters is the Aphex 230. I hope to someday use one of these in my recordings, but it was - and, at the moment, still is - beyond my budget. After purchasing the Heil PR40 microphone and Mackie Onyx Satellite interface - as well as necessary gear like cables, mic stand, etc. - I only had about $200 I could afford to spend on a channel strip. Needless to say, this means that I needed to settle on a unit that is far from top-of-the line, but there are nonetheless several decent channel strips for podcasters in this price range.
After looking at a number of options, I finally decided on the DBX 286A. It receives solid reviews for its price (see here), and it contained the processing features I needed most (see below).
One thing that many people like is that the 286A does not use the complicated settings for compression, etc., used by most audio equipment: instead, for most features, it uses settings that simply go from 1-10. If you are intimidated by needing to learn complicated settings and audio ratios, this may be a significant advantage for you. To be honest, however, I dislike these non-standard settings. If you are new to audio processing, you might prefer the simple settings of the 286A, but you will probably find as you become more knowledgeable and comfortable that you dislike the lack of precision control that these simple settings provide. Despite this significant limitation, however, I found the sound quality of the 286A to be my favorite of the channel strips available in my low price range.
I'll briefly go through my settings on the DBX 286A so that you can get a basic idea of what I choose to do in my specific recording circumstances - you might find that different settings work better in your circumstances. For example, you can listen to this episode of PodSquod, in which Mark Jensen quickly sets up a 286A to his taste (he likes a very processed sound, and therefore tends to use higher settings than I do).
The 286A contains an adequate mic preamp, but I vastly prefer the Onxy preamp in my Satellite interface, so I run an insert cable from the Mackie into the 286A. Because the dynamic Heil PR40 requires a lot of gain, I have the gain on the 286A set at +40db to acquire a solid signal for processing. I also have the Highpass filter on the 286A turned on, thereby filtering out low-frequency sounds like hum and rumbles; this reduces some of the noise from wind and traffic outside my window, and completely removes the sound of my computer.
Because I am very soft spoken, I do not need a lot of compression in my recording (compression reduces the dynamic range of an audio signal, and thus prevents the distortion - called clipping - that occurs when your audio overloads an amplifier); I prefer the sound boost that comes from simpy increasing the gain to the boost I could get from compression. I therefore have the 286A's drive (which determines how much the compressor reduces a signal) set at 3, and similarly have the density (which speeds up or slows down the amount of time it takes the compressor to increase or reduce compression) set at 3.
As you can tell from listening to GET WISDOM, my speech is very sibiliant (which means, essentially, that I place too much emphasis on "hissing" sounds like the letter S, and I also struggle with silibant-sounding mouth noise). I therefore set the 286A's de-esser Frequency at 8k, which is the upper-end of the normal range for de-essing. Setting it higher begins to make me sound as if I have a lisp ("Thaint Paul then went to Ephetheth"). I set the Threshold at a moderate 3.5.
The 286 combines its equalization into the Enhancer's two basic LF (bass) and HF (treble) controls. These provide a great deal of audio color, so you will want to use them sparingly to avoid causing your voice to sound unnatural. I have the LF set at 3 to provide a little depth to my voice, but have the HF set at 0.
The Expander/Gate controls the level of a signal by "opening" and "closing;" this enables you to filter out - to an extent - unwanted background noise. This, along with the de-esser, is the main reason I wanted a channel strip. You will probably want to use something like this - in either hardware or software - to remove sounds like traffic, wind, dogs barking outside, etc. (I also sometimes need to remove less common sounds, like a highly irritable burro, llama, and group of goats, but that's another matter). I have the Threshold set at -15dB for moderate attentuation, but keep the Expansion Ratio at just above 5:1 for gentle expansion. The key with the Expansion ratio is to have it match as much as possible the the Compressor Density: both my Expansion and Density are around 10:00.
Finally, I have my output volume set at -8. This keeps my level consistently high without clipping.
In my next article, I will look at the software I use to record and mix GET WISDOM.
Posted in News