1. Register NOW and become part of this fantastic knowledge base forum! This message will go away once you have registered.

Piercing Vocals

Discussion in 'Vocals' started by Mike Simmons, Apr 5, 2002.

  1. Mike Simmons

    Mike Simmons Active Member

    Hi, I'm dealing with a song where the vocalist is belting out some really piercing frequencies. I've never recorded a voice that hurts like this... rip your head off, shards of glass in your eyes kinda pain... the basic tracks are alright but by the time the track is compressed/L-1'd... ouch!

    Multiband compression at about 2k tames it okay but seems to have taken it's toll on the presence of the voice in a way that's not so good. I've tried pulling the offending sections to a new track but that just makes the difference in presence more noticable.

    Anybody have any ideas that might help pull the pain level back a bit?

    By the way the chain is: Blue/Dragonfly to DVC to PSX100 into PTmix at 24bit/44.1

  2. pan

    pan Guest

    Hmm, put a deesser in the chain, the Behringers do a good job at that...;-)...seriously! Waves make a good Plugin, to deal with it later on.
  3. Lethargy01

    Lethargy01 Guest

    Heh, kinda funny you brought this up. I had a similar situation about a week ago. Just a really high pitched, whiney, horrid singer I had to deal with. You know, the kind that makes the dogs run for cover. :) Anyway, I started with a Dragonfly with him, and it was just unbearable. That mic has a lot of top end in my opinion, and just really seemed to bring out that quality in this persons voice. So perhaps changing to a different, more neutral, or even more bottom end sounding mic would help you out a bit. Worked for me. Kept the really high things a bit more under control. Still sounds bad, but now it's just because his notes are all off, not because of too much high end. :) lol. Oh well, everyone has to deal with it sometimes I guess....

  4. pan

    pan Guest

    Ah you were talking about that Whitesnake-Symptom of Metal singers. A feedback destroyer could help :D :D :D

  5. anonymous

    anonymous Guests

    Since you have ProTools, why not just automate an EQ that cuts out some of the offending frequencies just on the notes that require it?
  6. CyanJaguar

    CyanJaguar Member

    why am I not surprised that a BLUE mic was used?

    the blueberry reviews that I read before buying said that the mic was optimised for vocals. Big untruth. They should have come out and said that the mic is optimised for SOME male rock vocals ONLY

    I tried it on a femle opera singer. Sorry had to switch

    tried it on a male crooner. No workie

    tried it on rap vox. Totally short of the mark
  7. Mike Simmons

    Mike Simmons Active Member

    Thanks for the suggestions. I'm guessing that the de-esser will work like the wave c4. So I'll give that a try and see if that helps.

    I probably shoud have put a 57/58 in front of the guy but I'd just pulled him off a Brauner Valvet which really didn't flatter him, he was nervous/self-conscious and I wanted to keep things moving. It was sounding fine during tracking. While my dragonfly is certainly very colored, it does a nice job in a lot of applications.

    The feedback destroyer... joke? I guess it's all hanging out in that 1k -3k range... (I don't have one though).

    Automated EQ sounds like a plan. Any tips on how to set this up quickly?? I'll give it a try! My best EQ plugs are Focusrite and Rennaisance. This is probably my best bet though. I know there was a thread on automating plugs not too long ago but I have to admit to being kinda slow at this.

    Thanks for the tips, they should keep me busy for a couple hours!!
  8. anonymous

    anonymous Guests

    This may not be "quick", but it probably will work. Add your EQ plug-in to the vocal track and select a particularly offensive section of the vocal. Boost one band at least 6 dB and sweep it up and down manually while playing the track until you find the spot that makes the vocal sound the worst. (Or if you have something like PAZ analyzer you can look at a frequency snapshot and note where the biggest spikes are). Once you pinpoint the offending frequency, play with the amount of dB of cut, as well as the Q to try and get the best compromise of smoothing out the vocal sound without making it sound too dull or otherwise unmusical. Chances are, what works on one offending section should work pretty well on other similar sections.

    Now for the automation: you could try automating the mute and unmute of the EQ, or activating/deactivating that particular EQ band - but often that creates audible pops or clicks. I prefer to automate the band gain - then you can draw in a slanted line on the way in and out of each section, which will be far less noticeable.

    This is a brute force method which requires manually locating and automating each offending note. I use it when there's only a few such notes in a song. But if there's tons of them and you're looking for something less labor intensive, you might try using a multiband compressor with a sidechain. Copy the vocals to a 2nd track - and cut everything but the "bad" frequency - use that track for the sidechain only (don't output it into the mix!) that will trigger the multiband compressor - set to compress only around the offending frequencies.

    Hope this helps!
  9. planet red

    planet red Active Member

    Or even easier, set up the eq on the channel and find the offensive frequencies, and audiosuite it on the offensive parts. If it doesnt work you can always go back and get the original track back. I normally just duplicate the track and play with it for awhile and if it doesnt work i still have the original and dont have to go looking for it. I like audiosuiting gain changes and eq's a lot more then trying to automate stuff. And doing gain changes on esss's sounds a lot better to me then just using a deesser.

    Either way I have a fatso coming monday so hopefully that should be able to soften some of the harsh/high pitched singers i work with.
  10. wiggy

    wiggy Guest

    Hey perhaps u could try a humble SM57/58....

    Too many people dismiss them as good vocal mikes.

    ride the compression on them and have a nice pre that can give them lots a juice. I have had great vocal cut with either 57/58 into a neve 1073/1272 and some simple compression from a RNC.

    If its not bright enough u can allways put an eq in line to ad or subtract a few nasties or some 'fairy dust'.

    Have been curious to try the Blue mics... but will be cautious when adutioning soon.

    Hope it helps!

  11. anonymous

    anonymous Guests

    Focusrite ISA series is fantastic at whipping out 'telephone' frequency at very narrow bandwidth but leaving the signal in relatively good shape... You get 2 x chances then a bass boost / cut (with HP filter off too) & finaly a super top HF boost / cut.

    It's THE tool for the job, once mic & pre selection has left you STILL with a piercing vox,

    Allso try adding bass, to leave the mid less at the forefront..


  12. tertsipede

    tertsipede Guest

    This is a very familiar situation to me as well!

    All of the suggestions above sound very good, but do not forget simple volume automation. After lots of compression and limiting some parts just sound too loud, if not automated carefully.

    Good Luck!
    - Teuvo
  13. RecorderMan

    RecorderMan Distinguished Member

  14. Mike Simmons

    Mike Simmons Active Member

    I just wanted to say thanks to everyone who responded. By the way, this worked really well for me:
    "you might try using a multiband compressor with a sidechain. Copy the vocals to a 2nd track - and cut everything but the "bad" frequency - use that track for the sidechain only (don't output it into the mix!) that will trigger the multiband compressor - set to compress only around the offending frequencies."
    Thank you Littledog!

    Finding the worse parts was made way easier by monitoring the sidechain. (duh!) I used a de-esser as the trigger and the C4 on the track.

    I'll also keep the SM7, Royer-121and ISA on my wish list.
  15. future/now.

    future/now. Guest

    should of switched the mic when you were just gettin' started, but now... slap a tube across it & if you can get your hands on a EL8 Distrssor- the band emphasis detector mode, 10:1 ratio does the trick everytime. (automating in PT sounds pretty silly)

Share This Page