From the May 1989 issue of Radio And Production

The Voice & The EQ

by Jerry Vigil

The equalizer is probably the processor most commonly used on the voice. It is also one of the most misused processors. Without an "ear" for EQ-ing, the end result can be worse than no EQ at all.

Let's get some EQ facts out of the way before we go on: 1) You must have decent monitors to hear properly the effects of your adjustments. 2) Boosting or cutting frequencies below 100-150Hz or above 11-12khz usually won't affect the voice. 3) All voices have different frequency characteristics, thus requiring different EQ settings to get a similar effect. 4) Boosting bass doesn't make the voice deeper. (A pitch shifter does that.) Boosting the bass tends to give a voice more "warmth" instead. 5) The type of mike you are using will affect how you set EQ. 6) How you work the mike affects how you set the EQ. 7) How your station processes its signal between the control room and the transmitter affects your EQ-ing efforts in production. 8) Different sexes and age groups have different "tastes" for EQ. Some demo-graphic groups prefer a lot of bass while others favor rich highs. 9) If you have several years of jocking with loud head-sets, your ability to hear high frequencies may be hampered somewhat.

Considering all these facts, if you can EQ a voice in the studio and hear on the air what you wanted to hear, you have an excellent "ear" for EQ-ing. On the other hand, if your ears are a little out of tune, a few exercises in equalization might help you achieve that flair for EQ-ing by ear. The only substitute that comes close to a good ear is a spectrum analyzer, but it's unlikely you'll ever convince management to buy one so you can use your equalizer.

Training your "ear" may or may not work for you. It may be that a person either has the "ear", or they don't. The following exercise will either help you, or it won't do a thing; but it's worth a try if you'd like to improve your ability to EQ by ear.

Start by finding a voice you can use as a reference. If you voice spots or promos, you're probably most familiar with the sound of your own voice, so use it. Otherwise, use the person on staff who does the most voice work.

Record a 60 second spot, voice only, to 2-track using your reference voice. Now find a piece of music that you think represents the average piece of production music in your library. An instrumental track from an album will work just as well. Just make sure the piece you pick doesn't have an abundance of bass, mid-range, or high frequencies. Now run the voice track through your equalizer and play the music underneath. Get a good mix of levels. Have your EQ set flat to begin with, then start making your adjustments. Leave the low frequencies alone for now, just adjust the mid-range and high frequencies. What you want to try to do is adjust the high frequencies (around 10kHz) on the voice track so the "S" sounds stand out as much as the cymbals and other high end sounds in the music. Boost the 10k range as far as your equalizer will go, then bring it back to zero to get an idea of the range of your equalizer. Then boost it again while listening to the music until the high frequencies of both the music and the voice sound the same. When you feel good about that adjustment, go to the mid-range EQ adjustments (around 5kHz). Adjust them so the voice easily punches through the brass, percussion, and lead instruments in the music. Again, start by testing the range of your equalizer to get a good feel for it. Once you've EQ'd the voice so it sounds like the EQ of the music, stop the music and rewind the voice track. Listen to it without the music, cutting the EQ in and out to hear the difference. Initially, you may think there is too much high and mid-range boost, but chances are you are very close to the settings you need to use for any voice track that will be mixed with music. Transfer this voice track, with EQ, to cart and keep it as a reference.

Now get voice tracks from everyone on your staff that would ever voice a spot or promo. Put them all on one reel without any processing of any kind. Get your EQ'd reference voice ready for playback. Play the reference voice and then play the first voice on the reel, adjusting EQ until the two sound as much alike as possible. (Again, don't make adjustments to the low end, just the mids and highs.) Note the settings on your equalizer and write them down. Now set them all flat again and go to the next track and do the same thing, but don't look at or consider the settings for the previous voice. Make a note of the settings again. When you're through, you'll notice that you made different adjustments to each voice. This is because the voices have different frequency characteristics. Notice whose voice had more high boost. Notice who had more mid-range boost. Notice that boosting mids and highs is what you will be doing most of the time. If the person whose voice you're EQ-ing works the mike real close, you might even find yourself cutting some of the low frequencies to compensate for this. Remember that cutting low frequencies can have the same effect as boosting the other frequencies.

Refer to your notes each time you get a voice track from someone. Listen to your reference cart before making adjustments to a voice track. Eventually, you'll learn the approximate settings for each voice and your ear will begin to hear the differences between different voices as you EQ each of them.

As far as the low end frequencies go, when you're recording a voice track that, to your ear, lacks low frequencies, have the person work the mike a little closer. If it gets to a point where the talent is popping P's, then have them back off and boost the low end with the equalizer. Be very conservative here as it is very easy to turn a good voice track into a muddy one by the time it hits the air.

Some voices won't require any EQ to sound right in a music mix. Other voices may require a very high mid-range boost to get them where they need to be. Many female voices are this way.

Listen to your station for the spots you've produced with these new EQ settings. Do they sound right? Does the EQ compare to agency spots on your station? If there is an obvious splattering of highs or way too much mid-range, you may need to go back to step one (setting EQ on reference voice) and try again. Pick a different piece of music this time.
If you're producing a spot that is voice over music, do not EQ the voice until you can hear it mixed with the music. If you're just getting voice tracks, record them flat. Later, when you add music, adjust your EQ settings to match the EQ of the music.

If you're producing a spot that is dry voice only, you can back off on the usual settings you have for mixing the voice with music.

Last month, a subscriber wrote and asked if there was some way that he could just set his EQ and forget it. Obviously, with the different characteristics of each voice at your station, this is unlikely. However, there are processors on the market that can eliminate your EQ problems. There are a number of digital processors with parametric equalizers that would be perfect for the person who doesn't want to mess with different EQ settings for each voice. With these processors, you can set EQ for each voice on your staff and store them in memory locations. Next time your mid-day jock walks in to lay a voice-track, you just recall that EQ setting and you're set to go. Orban's 787A Mic Processor was designed just for this purpose. It includes EQ, compression, and a de-esser all in one, with room to store different settings for 99 jocks! So yes, there is a way to "set it and forget it".

The best way to develop an "ear" for EQ-ing is to play with the EQ a lot. Listen to agency spots and try to match their EQ. Get to know your equalizer well. Eventually, you'll be making adjustments to your equalizer for a voice track before you ever hear the track.

In a future issue we'll look more closely at graphic equalizers and parametric equalizers and discuss their differences as well as advantages of one over the other.