Removing The Boxy Sound From A Voice Recording

it's really common to have a boxy sound when you're recording your voice it really all depends on the room that you're in your recording space but you could have a an acoustically treated room and still have that boxy sound maybe you're in one of those vocal booths that could be quite expensive or one that you've built yourself still you can't get rid of that boxy sound to your recordings well i'm going to show you how to deal with that with a parametric eq i'm lenny b voice over audio engineer let's make your voice sound better this voiceover audiophile submission is from george kate he is uh from chicago and he's a voice actor using a studio brix one vo edition uh home booth and if you don't know what those are i'll put the link in the description so you can check them out but uh he is using a sennheiser mkh 416 into a focusrite scarlett 2i2 he's using cubase as his daw and um he has fabfilter pro-q3 which is what we're looking at here but what i want to focus on for this video is eq'ing the this room now it's a professional voice over booth he has a pretty good sound but still there's a type of a boxy sound he's experiencing and this is common with smaller rooms that you're recording in voice over spaces so even if it's professionally treated it's a professional room even if you have bass traps and uh and other types of absorbing material in the room you can still need to even and balance out the frequencies and that's what we're going to do and you're going to see the difference between the before and after we balance them and why it's so helpful and how good the end result can sound so first we're going to listen to what we've got to work with here let's take a listen to his raw file there's no compression no eq this is just his mic the mkh 416 straight to the daw it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary news sources are becoming more popular the first thing i notice is there's very minimal audible reflections especially in the high frequencies it sounds pretty dead but i do hear a type of a boxy kind of a sound and that's the low mid frequencies and and here's what happens just a little bit of physics and background so you kind of understand a lot of the absorbing panels are physically thinner they're one or two inches thick and once you start getting in the range of the low mid and low frequencies those waves those audio waves and pressure the waves are physically bigger and they pass right through that thin material so the only way you can absorb the lower frequencies is with physically thicker absorbent material or scientifically designed absorbers that have they've got membranes in them that actually catch some of that lower energy in the sound pressure waves so it's common to have this type of problem in a smaller room and it's all based on the length width and height of the recording space that you're in so i'm not surprised by this but there's a way to kind of even out the frequencies it's not a it doesn't destroy the recording but that's where eq comes in to correct the imbalance so let's listen again and we'll look at right here i can hear him and i'm anticipating that they're in this area here it's no secret that a lot of legacy news organizations are struggling all right this is i kind of think it's around this area let's see if we can identify it we'll raise the specific band and we'll sweep through the frequencies and once i identify the loudest one the one that's louder than all the others that's where i'm going to work on it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary news sources are becoming more popular i actually hear another one just a little bit higher so i think there's two here it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary news sources are becoming more popular readers are growing frustrated with paywalls one of the things really cool about the um fabfilter pro q3 is you could change the slope of uh the band that you're working with and i think what i want to do is see if we can get something a little wider here but then not as deep in it in a specific area it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary news sources are becoming more popular all right to me right away i can hear that the boxy sound is minimized and you know like i said i anticipated it being in this area and it was and so now let's take care of some of the very low material and another great feature we can use in the pro q3 is the solo button so i can actually just hear uh everything to the right of this uh of this band while and while i'm sweeping through and what i want to do is i want to find just the absolute lowest material coming from his voice and then everything to the left of this band we're going to remove and filter out it's no news organization [Music] but now now that may be difficult to hear but um if you have a subwoofer in your monitor system or you're listening in earphones you might be able to hear it so everything uh lower than 54 hertz is going to be filtered out but i still want to make sure that i keep that low material because i want to use it later down the production and the processing path i'll play it again and you'll see that there's still material it's being filtered out but it registers still on the frequency analyzer here you can see the material it could be room noise it could be just really low frequencies it could be explosives it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers so it's this stuff right here and you know it may not be totally apparent but one of the things you need to consider is this information may affect the compressor that we are going to use later down the processing path it actually would work on the detector circuit of the compressor if you don't understand that don't worry about it right now but just know that we don't need that we don't want it to affect anything we do later but i do want the lowest parts of his voice so i have a full spectrum picture of his voice that i could sculpt and manage and mold with eq and compression down the path later first we're just concentrating on eq and balancing this out so let's continue and see if there are other uh areas that we can address it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers so the base part of his voice 140 around there is prevalent at certain times and i want to use something really cool that the the pro q3 does and some other dynamic parametric eqs and we're going to add a compressor and make this band dynamic so it only reduces once it passes a certain threshold only on that band it's uh really really cool it gets pretty deep uh as far as the concepts go but just follow along here so you can see what's happening it's no secret first i'm going to identify that band that is much louder than the others only on certain words it's no secret that a lot of legacy news organizations are struggling you can hear news he hits this tone here but organizations there's nothing listen again it's no secret that a lot of legacy news organizations are struggling so when this frequency band is apparent and loud i want to be able to attenuate it but when it's not being used i don't want to have any material being pushed down in this area because i still want some of that base so if we were to just lower it it would take everything out and lower everything in that range but i only want it to be lowered at certain times so we we make this band dynamic and then you've got a threshold here and so what happens is the threshold is available you can see it right here and set the threshold so where this dynamic on this band it dynamically reduces information in this area only when it passes a certain threshold that you can set it's no secret that a lot of legacy news organizations are struggling it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary news sources are becoming more popular readers are growing frustrated with paywalls and the use of ad blockers is increasing it's no secret that a lot of legacy news organized all right again my goal here is to balance everything out and make everything kind of that wall of sound and every frequency so let's continue and see if we have some other bands there's a couple up here that i want to work on it's no secret that a lot of legacy news organizations i'm going to split this high end you can see that there's a lot more energy in this high end at certain words we're going to make these dynamic again but i'm going to split this area into two sections and this is kind of how i approach this the siblings the s's and that sibilance that usually a de-esser may work on is kind of in this area right here so i'm going to separate that as one thing that i'm going to to process independently from just the high information and the even higher in the error and the high frequencies and i'll show you how i do that again i'm going to choose a band and identify it and because i know um that there is a very specific band uh for the sibling sounds i want to try to identify those and just isolate those first it's no secret that a lot of legacy news organizations so i think that's probably at about right there so i'm gonna do the same thing i'm going to lower them just to touch and we're going to make them a dynamic so this band specifically is going to be attenuated kind of like it's what that's exactly what a de-esser does but i want to make this as even as possible first and we may use another de-esser later down the path there's no rules you kind of use the tools like eq and compression uh to do and shape the sound and it's art you know you uh if i process a voice today and then i have a clean slate and and make everything uh the default settings and process it again tomorrow it is going to they'll both sound good but it may be done a little differently it won't be exactly the same because you use these tools to create a sound that is pleasing to you there's no right or wrong answer it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary new sources are so you can see that's just softening the sibilance a little bit um it's a 3 db reduction but only at certain times you can see that the dynamic band of the cq is attenuating that area only when it's necessary only when it raises above the threshold that i've set it's no secret that a lot of legacy news organizations are struggling and this ener the energy in this area is louder when he has his t's and sh sounds so i think that uh smooths things out a little bit and now we're going to work on this higher band that also has quite a bit of energy and um i'm using these slopes a lot more than ever before i really like them for voice it's no secret that a lot of legacy knew now secondary news sources are what this allows you to do is you know you can um attenuate and work on and kind of shape and mold the dynamics of bins completely independently because i still want this material i actually prefer a little bit of high end i like it i think it crisps things up uh and makes the voice more intelligible and understandable and also um you know that's the most important thing to make an emotional connection to whoever you're speaking to or whatever you're marketing or whatever the story that you're telling so i think it's really important so i don't want to lose it but i just want to minimize it when it goes overboard so we'll do that again for this area it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary news sources are becoming more popular readers are growing frustrated with paywalls and the use of adblockers is increasing it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers the whole concept is to try to balance things out to get that full wall of sound so it's even from every frequency space and i will say that you know there's the boxiness that i corrected this is actually eq'ing and it's not dynamic i just wanted to remove those because they were apparent in every single uh phrase and word and vowel and consonant that he was saying these frequencies were resonating in his uh vocal booth based on the length width and height of the room and one more thing before we discuss the in and output levels i want to put a shelf on just the low energy just want to lower everything in the bottom end a little bit and i'll explain why here it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary news sources are becoming more popular readers are growing frustrated with paywalls and the use of ad blockers is increasing it's no secret that a lot of legacy all right so we just overall uh reduced some of the lower energy and um one of the things that takes time and experience i've done this so many times is i'm anticipating with the compressor or multiple compressors may be doing down the signal chain i just know from doing it so many times that i just want a little bit lower energy in this very low end or really the lower half of his voice and it's not much i mean we're talking about 1.5 2 db 1.6 db of reduction but it makes a difference down down the path all right now i want to turn your attention to the in and output meters so this level meter uh loudness meter is by you lean one of my favorites uh it's actually free i'm using the pro version here which comes with just a few other features but you get most of them with the free version everything you need with the free version you lean.com i'll put that link in the description the audio we're hearing is first going through this first u-lean meter so we're getting a level on what the original recording was and and uh the loudness of the original recording then it's getting passed to the um fabfilter pro q3 and the work that i've been doing and then it goes to the second eulene meter and the reason for this is i want to know what kind of uh overall loudness effect that this eq work has done and i want to balance it out and this allows me to do them when they're back to back like this what i'm going to do is use the output volume here and this is post eq changes that i've made and i'm going to raise it so these are matched and then we'll listen to the before and after and see what kind of positive changes we've made to the overall sound it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers it's no secret that a lot of legacy news organizations are struggling all right we're about the same here you know it's uh if you're 1db off that's okay i mean i like to get as close as i can but the overall integrated loudness which is an average um is very very similar just a couple points off so what what that means is i've removed this low energy i've removed energy in some of these frequency bands this is all subtractive eq and if you add it all up it's taking about 3 db you see we've increased the loudness here 3db going out of the eq and it's makeup we're making up the gain and and bringing it back to the level so then when you compare before and after the eq sounds you're really getting a true representation of the work that you've done because it's the exact same level or average loudness and you can just hear just what the eq is doing now i receive submissions uh audio recordings from really world-class voices that do movies and audio books and video games and animation and explainer videos and youtubers however i also work with beginners and i'm happy to help you if even if you're simply beginning to record your voice for any type of project and here's how to do it first thing you do go to my website it's lennyb.com and you'll see the red button up at the top there just click that button and it's going to bring you to uh some instructions and it's just how to organize your audio file in the way that i need it uh at least 30 seconds doesn't have to be too long though i don't need longer than two minutes it's not necessary but i do need unprocessed audio i want your microphone directly to your interface and right to your recording no compression no eq because this like this way i can get a complete picture of what you are up against and how we can resolve issues or just simply optimize the audio make sure you get your recording level right uh you can include five to seven seconds of room tone just so i hear what kind of noise you're dealing with i'm looking for the wave format and if you don't have copy there's a link where you can get some copy that i've provided for you and then make sure you title your file with your first and last name so really simple stuff it's easy to do then there's a little form to fill out which gives me some information about the equipment that you're using you know where you're from and what kind of interface you have what platform mac or pc and also a place where you can describe um you know what kind of acoustic treatment you're using this way when i have a good picture of what you're working with i can help you even further then you're going to get a link and i email you a link where you could just drop your file in dropbox and it goes right to me let's reference the unprocessed version of this clip without the eq and then we'll switch it on and hear the differences it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary news sources are becoming more popular readers are growing frustrated with paywalls and the use of ad blockers is increasing sounds uh more even more balanced i may do a little bit of um fine-tuning here it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary news sources are becoming more popular readers are growing frustrated with paywalls and the use of ad blockers is increasing it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary news sources are becoming more popular readers are growing frustrated with paywalls and the use of ad blockers is increasing it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers all right i've done a little bit more work here and i think maybe if you were following along i've widened this particular uh band with the slope and then added something a little bit specific because i just heard that still heard that uh one frequency band we've made it a very surgical thin cue and added another one let's listen again it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary news sources are becoming more popular readers are growing frustrated with paywalls and the use of ad blockers is increasing so i feel like we've removed that boxy sound when you look at this whole eq you know the bands that were attenuating are either dynamic so they're only working at certain times and we're talking about three db of uh attenuation it's not much it doesn't you don't need much because i want to balance it out and i'm using the information the data from the spectrum analyzer i'm using my ears and i'm making a decision based on what i hear i'll point out that the the changes we've made to these bands um the dynamic bands that are only working when they're needed and also uh to remove this boxy energy and the low frequency energy these are all um you know most of them are three db reduction less maybe 2 db and then the only one that i'm actually taking down to 9 db which is quite a bit is that particular boxy resonant frequency but i just want you to make note that you know you don't have to eq much i know there's a lot of bands here and a lot of things going on but the actual reduction in volume in those specific bands is is not much 3db or or less for most of these now that we've got the eq taken care of we're going to take a look at the next step in the processing and for this particular voiceover again i'm using one of my favorite multi-processing plug-ins the sheps omni channel by waves and i'm just going to quickly walk through this i think we'll we'll go through some of the details in another video we've added a little saturation um i'm using uh expansion on this um instead of a gate it's the gate expander but we're closing down the audio just a little bit to remove some of the the noise in the background i'm doing some sweetening with eq when we look at the other eq this is subtractive eq and most of the eq that we're doing in the sheps omni channel uh most of it is um additive so it's sweetening some of the pleasant frequencies there is one i am reducing a little bit of that low low mid energy still because uh where the boxiness was uh you'll see that i put the eq in front of the de-esser and i am reducing a little bit more of the sibilance here then we're using a transparent vca compressor uh this is a rather fast attack and release at a quite uh i'd say it's higher than most 3.5 um ratio and we're just taking about one to two to three db uh reduction in gain just from the transient so i'm really using this to take off the highest peaks of the loud parts and then i've got a second uh compressor which is an opto compressor which runs much smoother and slower at a lower ratio to just even the dynamics out further and it gives the whole thing a nice sound okay let's listen to the entire clip before any processing that we've done here with the pro q3 and the sheps omni channel and then after and see what kind of a difference uh this processing made especially by first uh evening and balancing those frequencies with the eq it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary news sources are becoming more popular readers are growing frustrated with pay walls and the use of ad blockers is increasing you can see that we've removed most of that boxiness and uh the sound is more upfront uh the entire uh frequency space is represented it's a nice big a big thick wide frequency representation and again i've said this over and over again but really what you want to do the entire goal for this whole thing is to make the most strongest biggest emotional connection you can with your voice and these tools allow you to bring that voice really up front and every voice is different every room is different every microphone is different so you need to know how to use the tools to do what you want to shape that to shape that audio that recorded voice and and bring it up front to make that emotional connection it's no secret that a lot of legacy news organizations are struggling they've always competed for viewers but now secondary news sources are becoming more popular readers are growing frustrated with paywalls and the use of ad blockers is increasing lenny b.com is the website and get more information about my processing service training courses and i've also got free voice over tips and techniques i invite you to subscribe to the youtube channel for more voice over audio videos like this one and i love to answer questions and i really do welcome your comments when you're voice acting when you are creating something when you are putting your voice underneath some video when you're doing a presentation when you're doing an audio book anything when you record your voice that sound that you hear should inspire you to want to record and do better and emote and and and tell your story and if that's not happening let's make your voice sound better thanks for watching

Channel: Lenny B

Views: 26,251

Keywords: Voice Over, Voiceover, Microphone, Training, EQ, Scheps Omni Channel, Fabfilter Pro Q-3, Voice EQ, Bill Dewees, Both Junkie, A VO's Journey, Tom Buck, Dan Worrall

Id: T9g7bpOJ4l4

Length: 25min 36sec (1536 seconds)

Published: Tue Jun 14 2022