HELP! Minor Disaster

GroupDIY Audio Forum

Help Support GroupDIY Audio Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Lalalala

Member
Joined
Jun 5, 2004
Messages
18
Hello All,

A thorny problem that happened tonight. Would appreciate your ideas and expertise.

What happened:


I was fiddling around with my electric guitar plugged into a smokey amp. I had my 16 Track Korg Digital recording on a single channel with a mic plugged into the channel. Inspired, I recorded a series of 8 songs--vocals and music-- straight into the mic.

Stopped. Listened. The amp overpowers the vocals--you can hear the vocal throughout but can only sometimes hear them loud enough to make out the words. I wrote them as I was singing them, so can no longer remember what they are--but I like them, and would like to be able to hear them so I can write them down and rerecord the songs properly.

So, I've been fiddling around with the eq on this thing--I'm a relative novice--trying to bring up the vocals. This has some useful effect but not quite enough to bring out the vocals, although I don't think I have exhausted all the possibilities. It has an eq that you can apply per channel as an insert effect, as well as an eq that you can use on the channel as a master effect. I've also been toying with other effects--exciter, reverb--trying to bring out the vocals. I am extremely new to eq and effects in general

I have a tenor voice. Any clever ideas on how to bring the
vocals or subdue the guitar on this thing so I can hear the vocals? I just need to be able to hear them well enough to distinguish the words. Your creative ideas are appreciated!

Best,

Alan


Dr. Alan J. Lipman
Washington, DC
 
uh-er-uh. The guitar is probably in the same frequencies as the vocals, so eq won't help much. My advice is to listen closely to the tracks while they are still fresh and pull what you can. Next time record with two tracks-you got 16!

Joel
 
hey alan-

if all you are trying to do is deciper and write down your lyrics, apply a hipass filter and adjust the frequency untill you cut enough of the guitar to hear the vocal. If you have an overly buzzy guitar you may be out of luck. Start with a hipass around 2K and see what you have. It really all depends on what the frequency response of the mic is and how it reacted to your vocals.

dave
 
pull back frequenties in the 250-400 hz region, there's a lot of information what can be lost there. As soon as you get some headroom back, try boosting from 850 up to 2500 and see if this makes it better comprehensible. In case it get's worse, cut there too of course...

In case this fails... post mp3's so we can hear the problems as they occured.

good luck,

Tony
 
Use the techniques above to try and get rid of the vocal
Once you have got rid of the vocal copy vocal-less track to another track
Phase reverse the new track
Play back both tracks - the guitar should disappear

Not much chance of it working - but have a try
 
> The amp overpowers the vocals--you can hear the vocal throughout but can only sometimes hear them loud enough to make out the words.

All music is based on the voice. The guitar covers all vocal ranges from baritone to alto, and often plays through your tenor range.

But words are a mix of vowels (a,e,i,o,u) and consonants (s,t,m,p,g,etc). In speech the vowels are stronger than consonants; in singing we "sing" the vowels and separate them with consonants. But the consonants carry the intelligibility: you cn wrt jst th cnsnts and still figure out the words, some languages are written without vowel-letters. The guitar's "consonant" is the pluck-noise, the lips have a much richer range of consonants.

Many (but not all) the critical consonants are mostly high-frequency noises. You may have to turn to a stronger EQ than commonly used in music. (However, if you have a "telephone effect" filter, that's likely to help.) Cut everything below 1KHz as hard as you can. I would use at least 18dB/octave (I'd never use that steep a filter in the midrange of music). You may need to cut as high as 3KHz to reduce guitar overtones and bring out the vocal spit and pop. It is going to sound like crap, not like voice, and you may have to listen to it over and over to train your ear to the badly whacked sound. It might help to listen to some other tracks, that you know the words of, with the same filter, to train your ear.

Cutting everything above 7KHz may also help, since that is mostly poorly controlled lip/tongue noise.

After you get most of the guitar and voice fundamentals cut down, you might try clipping. First apply a 6dB/oct rise from 2KHz to 8KHz, then over-amplify the signal about 10dB above "Normalized". Then play that with a treble-cut. You can clip the hell out of speech and still learn to understand it, especially if you boost the highs up front and then cut the highs after.

Try changing the pitch a little, a few semi-tones. That may throw the pitch-perspective enough to make some sounds stand out differently. Try shifting the filtered track down in speed and pitch an octave, and l.i.s.t.e.n v.e.r.y s.l.o.w a.n.d d.e.e.p.

And for future reference: the room-level of most guitar amps is higher than the singing-level of most singers. Yes, I know a few singers who can peak louder than a Fender Twin, but they can't hold it for a whole aria, not even a verse, while the Fender can (and usually does) cruise that level all night. Eat the mike. Or switch to a small-body soft-string UN-amplified acoustic guitar.
 

Latest posts

Back
Top