Jon Fairhurst
December 29th, 2007, 01:11 PM
Meryem,
You bring up a good point. For people shopping on a budget, audition the mic if you can with your own headphones and your typical talent. Ty is right that with a good setup, EQ and compression can make a world of difference. Having a mic that sounds good out of the box and doesn't have *too much* character really helps though.
Here's a good VO trick. Copy your recording to two tracks. Leave one uncompressed. Compress the snot out of the other (for instance 20:1 @-20dB.) Now mix to taste.
The uncompressed track keeps the peaks, so the result still sounds lively and crisp. The compressed track pumps up soft vowels and makes the voice sound full.
Bus the two tracks through a common EQ. The fundamentals are in the 200-400 Hz range. Adjust this for the bass balance of the voice. If there are intelligibility problems, boost the frequencies around 1.2 kHz. That should help bring out the consonants. Vocal character is often around 2.4 kHz. If the voice is too nasaly or harsh, cut it a bit. If the voice is hollow or it's hard to distinguish between multiple speakers, try a boost. Finally, the 5k-15kHz range is where you find the sparkle or "air." Boost this to make a dull voice shine.
These are only rough guidelines. It really depends on the voice and the recording. But it's a good place to start.
Also, you can make sharp frequency cuts, but (pretty much) never make sharp boosts. Boosts should be wide and smooth and cover a range of frequencies.
I used the above approach on an amateur with a reasonably deep voice recently. When he heard the raw recording, he wasn't feeling too good about himself. After the processing, he was all smiles. :)
You bring up a good point. For people shopping on a budget, audition the mic if you can with your own headphones and your typical talent. Ty is right that with a good setup, EQ and compression can make a world of difference. Having a mic that sounds good out of the box and doesn't have *too much* character really helps though.
Here's a good VO trick. Copy your recording to two tracks. Leave one uncompressed. Compress the snot out of the other (for instance 20:1 @-20dB.) Now mix to taste.
The uncompressed track keeps the peaks, so the result still sounds lively and crisp. The compressed track pumps up soft vowels and makes the voice sound full.
Bus the two tracks through a common EQ. The fundamentals are in the 200-400 Hz range. Adjust this for the bass balance of the voice. If there are intelligibility problems, boost the frequencies around 1.2 kHz. That should help bring out the consonants. Vocal character is often around 2.4 kHz. If the voice is too nasaly or harsh, cut it a bit. If the voice is hollow or it's hard to distinguish between multiple speakers, try a boost. Finally, the 5k-15kHz range is where you find the sparkle or "air." Boost this to make a dull voice shine.
These are only rough guidelines. It really depends on the voice and the recording. But it's a good place to start.
Also, you can make sharp frequency cuts, but (pretty much) never make sharp boosts. Boosts should be wide and smooth and cover a range of frequencies.
I used the above approach on an amateur with a reasonably deep voice recently. When he heard the raw recording, he wasn't feeling too good about himself. After the processing, he was all smiles. :)