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

Searching for a selective EQ based on content (inverted Dynamic EQ or some similar)

Discussion in 'Recording' started by pcrecord, Feb 27, 2013.

  1. pcrecord

    pcrecord Don't you want the best recording like I do ? Well-Known Member

    Hi everyone,

    My first question to this remarkable Forum ;)

    This might have been discussed many times but since new tools are created everyday, I bet it isn't a bad Idea to ask

    I need to fix an over limited vocal track. The content is really good except some occasionnal words that lost brillance du to over limited action from a tube preamp. The female signer was signing normally to the sound check and did'nt move away from the mic when going to some very powerful parts of the song. The band insisted to keep that take and the signer can't come back to redo it...

    Don't get me wrong I was able to fix the track to a good result but It's still not perfect to my ears.

    I've use some extanders, exciters, dynamic EQ, Multiband compressors and Demo a lot of plugins. I found that none of what I tested could affect only the parts missing those fequencies. To some extent I'm able to fix it but never without affecting the rest of the track, so when the loud words are ok there's always too much on the rest of the song.

    Now before I put all those words on an other track and process them seperately or go in an automation frenzy, Is there any tool, that can add some hi frequencies, only when there's less of that frequency present. I mean like an auto EQ that analyse the materiel and when it has less hi frequencies, it adds it ?

    Sorry for the long post ;)
  2. Kurt Foster

    Kurt Foster Distinguished Member

    dynamic eq or frequency dependent compression.

    i once worked with an artist who sang in a timber much like Jackson Brown. there was a harshness in his vocal occasionally when he's really belt out higher notes and i found a favorite compressor that worked very well for his voice which was a UREI LA-22. it has a side chain built in the detector that you can dial up to trigger the comp at a desired frequency.

    if you can't get an LA-22 (They are rarer than hens teeth) you can just patch an outboard eq into the side-chain of your favorite vca type compressor and dial it up so the compressor is more sensitive to the offending frequencies. the valley people dyna-mite compressor works very well for this. you can get them used on EBay or as a plug in from soft tube.

    this trick is a real lifesaver helping you to not have to use extensive eq over a whole track, keeping the track full bodied and beefey / crisp most the time and clamping down on the yuckies when needed.
    audiokid likes this.
  3. hueseph

    hueseph Well-Known Member

    Might also want to apply some volume automation or just some old fashioned vocal fader riding.
  4. pcrecord

    pcrecord Don't you want the best recording like I do ? Well-Known Member

    Hueseph, I might be force to do automation but since the customer is on a budget, I'd like to find a tool.

    Kurt, I'm trying to make the opposite we normally do. Some nearly shouting words sound muddy and missing presence because of the limiter. So I want a software that detects when some frequencies get lower and boost them. Most tools I tested will act when certain frequencies get higher but not the other way around. I could put too much treble and compress it but the result does not sound right.
  5. hueseph

    hueseph Well-Known Member

    Maybe you're just thinking too in the box. I think Kurt is suggesting a frequency dependent limiter but more in the lines of a high pass filter rather than low pass. A sort of "esser" as opposed to a de-esser.
  6. Kurt Foster

    Kurt Foster Distinguished Member

    exactly. set the compressor to emphasis mids and highs by boosting the low mids on the side-chain eq.
  7. RemyRAD

    RemyRAD Well-Known Member

    What you are asking about is dynamic EQ. Programs such as even an old version of Adobe Audition, 1.0/1.5 and now the free version of 3.0, for PC's.

    The stock dynamics processing program in Adobe Audition has the ability to draw your compression/limiting/downward expansion/gating in the dynamics GUI. And then you can add whatever bandwidth and window in whatever which way you want to process it. So you could make an upward expander that would only be sensitive to working on the higher frequencies. Followed perhaps by the necessity to then add DS-ing? Since there is no one single way to from a single plug-in. You are performing recovery and restoration procedures here and they don't make an app for that. You are trying to undo a beginner's screwup even if it's your own. Are you sure you are just not fixating on something? Why not send me a copy of that vocal track since this sounds like it will require some kind of forensic audio skills? Sounds to me, you are pissing into the wind? Turn to the east and you won't get your feet wet, next time LOL.

    Spectral processing may also be in order? And where each part of the different spectral bands will be working on narrow bandwidth sections and also utilizing other techniques. This will require some actual audio engineering which is not a drive-through McDonald's plug-in with a single button. LOL. Don't even know what made you thought this should be an easy recovery? It's going to take time, experience and talent before one could even attempt to make this forensic repair job, work. And you didn't even bother to post an example of the problem in which you are having? This is another mommy I boo-boo my finger again description. Because proper use of EQ and dynamics was obviously botched. Because good compression and/or limiting doesn't really cause this to happen. However it can without some kind of rehearsal or microphone test and someone who does not know how to tweak a limiter. And along with your inept skills, your reputation and accountability are on the line. And none of us here want to see you blow your opportunities or inept reputation LOL. Stuff like this happens to everyone of us every so once in a while. Most of us though, have been left to our own devices to save our ass. And in my case, reputation. I've already saved too much of my ass as it's really really big.

    PM me if you'd like? And please accept my humble apologies as I do not speak any French in spite of the fact of my first name LOL. I had trouble enough just asking for a receipt in this NICE French city even with my Berlitz book in hand. I mean it's even labeled NICE on the map but I never really found out what the name of the city with was? I also went looking for Vincent van Gogh's ear in St. Remy but couldn't see his ear anywhere? So I came back to the states, I went driving northbound and stopped in St. Remy, New York, to see if I could find Remy anywhere? And he/she was nowhere to be found? And I was turned around at the Canadian border post 911 for some prior possession and DUI charges I had years earlier. And how could they turn away a person whose first name is Remy? Especially when I had spent over $10,000 the previous year in Montréal? I thought they liked my money? Doesn't that also helped to bolster the Canadian economy? I wasn't anymore drunk or stoned than any other Canadian I knew? And where U.S. Customs indicated they knew I'd spent a lot of money there the previous year and the government just wanted some more of it. Too expensive an admission fee. Something goofy in your constitution there giving your government the ability to try and convict people of previous crimes they have already paid back their debt to society for, years earlier. And they even said I would have to attend Canadian rehab... for something that happened 10 years prior to my last visit to Canada LOL. We both have stupid constitutions in dire need of 21st-century upgrades. I mean I think Napoleon has been gone for a while?

    Now all I have to do is program my English voice recognition program to spit out French from English words? Where is it that the verbs or nouns go? I can never get that straight. But then someone like myself isn't exactly considered quite straight? Probably because I'm an audio engineer that just happens to be a... [Imagine what you will in this box]?

    Dr. Minard knows it's my birthday today. And so does Dr. Brassard. Salute!
    Mx. Remy Ann David
  8. pcrecord

    pcrecord Don't you want the best recording like I do ? Well-Known Member

    OK ! I might have find exactly what I'm looking for !

    To again better explain my search, it was for a Dynamic EQ but with an inverted threshold option.
    So when the gain gets lower the EQing starts ! See all the dynamic EQ I tested could only respond to a positive threshold (more gain = more EQ addition or substraction)
    You're all very kind with your suggestions, but I needed the opposite tool : (Less gain = more EQ addition or substraction)

    I downloaded Brainworkx bx_dynEQ V2 demo and will test it tonight. Hopefully it'll take care of my inconsistent track where some words have enough Hi frequencies and some need more.

    I'll get back to you with the results.
    Thanks !
  9. pcrecord

    pcrecord Don't you want the best recording like I do ? Well-Known Member

    Could be a Beginner's screwup. I could have lied and say it's someone else's track to get your simpathy but I did'nt !! :wink: Anyway, Rubbing my nose in it won't help a bit lol ! :redface:

    As I said the lady was a baby cat and transform into a screaming lion in the middle of the song, it was a surprise. I've lower the gain of the preamp and we made a few perfect takes (sonic wise) after that one, but when the band members decided what to keep before mixing, it was that same track I shoud have deleted... S..T !!

    The track isn't so bad, the band members like it but I don't. How many times it happened to you. Either, off pitch off rithme or else. I feel my job is to fix problemes as much as possible. I hear the problems, they hear the feeling and vibe they like. If it does'nt sound right, I feel I can't let it go. My name is on the record after all. eh eh eh ! But often they don't want to pay extra studio time, so yeah, I'm looking for a quick fix.. a pluggin to help make it better and faster then complex editing (I'm able to do btw)

    Ok enough said.. I got a new test to make this evening and I'll get back here to tell if the result were ok
  10. hueseph

    hueseph Well-Known Member

    There have been some good solutions posted. Also, you can't add what isn't there. Subtractive eq/subtractive dynamic eq is going to work better for you. I can almost guarantee it. When you push those frequencies on the vocals any compression you have is just going to squash them again and you end up with the opposite of what you are trying to achieve. Of course this is just my opinion but I think you should at least try some of the suggestions before you go off saying they won't work.
  11. Kurt Foster

    Kurt Foster Distinguished Member

    i am sure what i suggested is the answer you are searching out.
    it all depends on what freqs you want to add. if you want more lows when it gets quiet the dial it up so the lows are compressed when it's loud.. there's your subtractive eq. i think you are focusing too much on the idea that you are compressing a specific freq range instead of looking at how compressing that specific freq range affects the rest of the spectrum.
  12. RemyRAD

    RemyRAD Well-Known Member

    Here is another situation of using dynamic EQ in the way in which you want it to work. It's called Dolby B. It does exactly what you want it to do. It raises the high frequencies during the soft sections and lowers the high frequencies, the louder things get. Other variations have been used like this with other Dolby devices. It was quite common a few years back to take a professional Dolby A, 361 noise reduction unit. A special controller was made to plug into the front faceplate (actually behind the removable faceplate). Because the Dolby A was a type of spectral noise reduction unit, it did the same thing that the Dolby B, did. But it's split it up across the full high fidelity bandwidth from 20-20,000 in four groups. And this dynamic spectral processing enabled both compression and downward expansion to happen, individually, differently, in each one of the four frequency bands. It was used daily on Tom Brokaw for NBC-TV's Nightly News to process his microphone. So while providing dynamic equalization it was also providing downward expansion which worked as a single ended noise reduction device, as well. It was not used as a tape noise reduction system, in that application.

    Dolby B, was the consumer version and it only worked in the single high frequency band mostly above 3 kHz and beyond. But those were hardware devices. What you want can still be accomplished in software. But you have to understand the graphical user interface and if it features the ability to frequency weight the detector. And then you have to draw what you want it to do in the graphical user interface. And that's how I do it in software today. This is all based upon threshold settings, the dynamics curve you draw and how you tweak your levels. Without a basic understanding of how to do this, you're screwed. It requires a certain amount of basic knowledge and if you don't have that knowledge you're not going to get what you want, it's that simple. We've all been up against similar audio problems and we've all dealt with it one way or the other. And this is the problem between a hobbyist engineer and the professional engineer. Without the proper knowledge you're not going to be able to fix this. There is no gizmo designed precisely for this screwup. They don't create presets and effects to screw things up or to unscrew things down. And sometimes this requires extremely laborious processes within software. You might have to correct just a note and not the whole damned track? And that's part of your audio engineering job to do. Whether it's in software, on a digital console or in an analog control room scenario. And you won't get it fixed if you don't know what you're doing and it appears you don't. Nobody said this was an easy thing to do nor did anybody say that becoming a good audio engineer is something you get out of a box of cracker Jack's. You don't even become a good engineer if you have a PhD in the Recording Arts & Sciences because I've run up across engineers like that also that have made beginner errors in their tracking. But they have a PhD! Yet they still don't know anything. Because they don't do it. They learn things in a book and passed some tests. This guide Dr. Michael Cherry PhD, with such a lousy engineer even though he was teaching it, today he is a lawyer. I guess he finally figured out he was no kind of engineer? He couldn't even understand how I could get a good mix in his lousy studio with a PA board (because of specifications were better than that of an MCI recording console. But that's not a recording console. That's a PA board.) And he told me I would have to use his Valley People Digital Limiter on the stereo bus if I was going to expect to get a good or usable mix. And I told him to patch the damn thing out it's a piece of crap! And he argued that with me. And I three hours time, I mixed a half a dozen songs, of which he could not believe how fabulously sounded. What's that say about his PhD in recording? You should've seen that awful control room my God! University of Maryland Baltimore County campus. A drop ceiling superstructure with no tiles. Speakers on top of big empty metal rack cabinets that resonated like EMT plates. A Crest PA board (because of its superior specifications blather nonsense). So I really don't think much of recording engineers with PhD's especially. And I told him he was a liar. I told him that no University offers a PhD in the Recording Arts & Sciences. Well he corrected me there.

    Apparently at the University of Maryland, where they only offer a Masters in the Recording Arts & Sciences... if a PhD course is not available? You can create your own PhD course. So who the hell did he give his final thesis to? A physicist that knows nothing about audio recording or audio in general? How fraudulent is that? He couldn't accomplish a mix without a bus limiter. Which actually means he did know how to mix with a damn to begin with. He needed that crutch. Mr. PhD Dr. Michael Cherry who ha? Honest to God... I could not believe this poor excuse for a control room from a PhD? He was just another wanna be with money to burn.

    Do not ask for him as a lawyer. I don't believe his lawyering skills will be any better than his music engineering skills weren't?

    Beware of those who think they know. I don't know anything. I just do it.
    Mx. Remy Ann David
  13. pcrecord

    pcrecord Don't you want the best recording like I do ? Well-Known Member

    Hey thanks for all the answers !

    I was able to do a good job on the track. The brainworks pluggin did'nt help it gives a too inconsistent dynamic response with the content I needed to work on.
    I finaly accepted the fact that what ever I'd do, I'd loose some dynamics (what I was trying to avoid with my search of a magic wand) :redface:

    At first, positive dynamic of a dynamic eq, while fixing the parts that needed it, produced too much Hi frequencies on the rest of the song.
    So I went with first a Deesser, than a dynamic eq but I combined positive for presence and negative beavior on bass and mids. By choosing the right frequency centers that help the voice sound much more natural.

    The customer was impress and I'm more confident about letting it out like this. I guess I'm now the only one knowing there was a probleme..

    Thanks to you all.
  14. RemyRAD

    RemyRAD Well-Known Member

    That's great that you figured out a good or acceptable workaround.

    One of the things I've modified on a couple of my hardware limiters, the 1176's was this high frequency boost, detector mod. It turns the 1176 into a high-frequency limiter. So you can have this incredible outrageous high-end response without things getting screechy or tinny sounding. And I've been doing that since 1978. It works particularly good with crappy PA consoles being used for recording purposes. And where the Japanese/Taiwanese transformers dull the sound with every pass through. So you're always cranking up more high-frequency boost to cut through the transformer mud. And then the signal gets screechy and unusable without that high-frequency limiting. And where that high-frequency limiting takes out that screechy tinny sound. Yet you still have a sparkling high-end instead of mud.

    This is also easily accomplished in software of many different varieties and types but not all have that capability. And then I also had to use the downward expansion to get rid of the otherwise nasty high-frequency noise that I was also boosting. Otherwise it sounds like you have a couple of snakes for a chorus LOL. And no one wants to sound like a snake in the grass except for perhaps, the death metal rockers?

    I like pigs that find truffles.
    Mx. Remy Ann David

Share This Page