Getting clean crisp audio with Audacity and a Blue Yeti?

I’ve been doing YouTube for a few months now and the one thing that constantly gets me is my audio quality. I currently have me mic attached to a boom arm with some foam covering it and at any given time it is about 10 inches to a foot or so away from my face. Yet my quality still isn’t the greatest. I’ve been trying to stay away from Noise reduction because it seems to make my voice sound weird. I’m not looking to have my voice sound like the songs of angels…more like have my voice sound crisp and clear…like all of the other big youtubers do.

I’m a lot louder than most people when I do my commentary and I just want to know some good ways to make my voice actually sound good in my recordings? Here is a sample of o ne of me recent let’s play audios for you guys…I want my recording to sound basically like I’m sitting next to someone talking not like I’m just some digital voice thats been pre recorded. Basically I’m looking for audio quality of Jacksepticeye, Markiplier and all of them who seem to know the ins and outs of everything. I’m not saying better audio is the key to success or I want my voice to be like them…I just want to quality they have.

This is my recording. I need a knew keyboard since this one is mechanical and loud as all hell but minus the keyboard…how can I fix the quality? I have my gain set at about 9 o’clock on the blue yeti. Computer microphone settings are at 85/100 instead of the standard 45-50 it was at normally.

Any help? I just want the best quality I can get my hands on or wrap my brain around

Your audio is clipping quite badly at times between 1:46 and 2:18. If it’s not clipping on the Yeti (check), then turn down the recording level on your computer. If necessary, move closer to the mic when you are not shouting.

The hum in the background sounds like it is mostly the sound of your computer fan.

So do you think instead of having my boom arm attached to the back of my desk I should attach it to the side for ease of moving? I ultimately want it to sound good without having to move the mic or myself basically at all

Not to double post but,

With me being a much louder person do you think it would be beneficial to turn the gain all the way down and mess with the volumes in post? I did a recording today with the gain at 9 o’clock and I used a Nyquist code I found via older threads and at first it sounded great but as I got louder I noticed I was having some bad echoes and I could hear myself breathing and everything…which is definitely what I want to steer away from

You don’t have to worry about the mount vibration at all if you use a shock mount. I made one out of plumbing supplies.

A proper blast filter is not made out of foam. That’s a sure recipe for muffled audio. It’s the tennis racket thing in this shot.

It won’t eliminate the pops, but it will suppress them enough to finish the job in post production — and it doesn’t affect the rest of the audio at all.

I was having some bad echoes

Bad echoes? What kind of headphones do you have?

You should adjust your position and the recording settings so the blue wave tops are about 0.5 or so and the sound meters pop up to about -6. You’re never going to hit this exactly, but that’s the goal.


You’re a little too free-form. You can only get expressive and loud until the sound channel overloads. Then it turns into rough and crunchy and there’s no way to fix it.

I hear room back there. You can help that by recording in a room with no smooth walls and bare floor, or use a directional microphone and get closer. But that means you’re going to have to deal with the theater a little better. Too low and you complete with the noise and too loud and it overloads.

Cellphones get away with it by very powerful auto voice processing. Most times that kind of processing doesn’t make a very good show.

I didn’t notice any bad muffling. Are you sure you’re not crisp enough?


I remember where I heard that before. Are you fighting with the recorder, or are you supposed to be having an argument with somebody as part of the show? Most times you hear or see something like that it’s carefully staged so it can be recorded. People are not really walking around or waving their arms. You can’t actually have an argument. That would be impossible to record.


My “pop filter” is a sock rubber banded to the microphone as of right now haha. The headphones I use currently are Audio Technica ATH M50’s…so I’m not sure if thats a good pair of phones to use?

My room is pretty large and definitely has smooth walls. I’ve been looking into soundproofing foam for at least the wall behind me but I want to possibly put up some kind of curtain directly behind me to see if that helps? I’m a complete beginner when it comes to audio recording…my first thought was buy an expensive microphone and I’d be set to record immediately…that is definitely not the case I have learned.

Most of my arguments or recordings that are very loud or in your face kind of commentary are mostly me just trying to pretend the game or character in said game can hear me. I try not to stage a lot as I’m not that great of an actor, yet lol

If you’d like I can take some pictures of my current setup? I just want to achieve clean recordings…I notice with the gain all the way down and I scream, it still always seems to max out on the soundwaves? With the gain at 9 o’clock it seems to do that when I simply start my intro. I have no idea how to effectively use the compressor to be honest. I downloaded that Chris’s compressor thing but it still wont help me if I dont know what I’m doing. I don’t know if it helps but the microphone is mounted on a boom arm approximately 3 inches to the left of my head and about 10 inches to a little over a foot and it’s turned so the reciever is facing my mouth. My computer is in the little cabinet that is built for it on my computer desk which sometimes amplifies the sounds of my computer thats for sure…I’m also recording through audacity with my headphones plugged into the Yeti, I don’t know if that makes a difference or not

Okay so after messing around today, I have found that I seem to hit those levels of .5 a lot easier with my gain turned down more than the 9 o’clock setting. This may be a newbie question but is there a “perfect” Gain setting for these microphones or is it all based on what you are doing and how loud you will be?


The killer is “clipping”. If the recording is badly clipped there is no recovery. I presume that the Yeti has a clipping indicator of some sort (I don’t have a Yeti) - probably a red LED that comes on when the level is too high? To set the gain on the mic, shout as loud as you will in your recording and adjust the mic gain level until the LED just comes on, then back it off a little so that it does not come on. Then set the recording level for your computer. Should as loud as you will, and adjust the recording level so that the waveform in Audacity is about half the track height (around -6 dB). Your recording will probably sound a bit quiet when you are talking normally, but if you want to shout in your recording then you will probably need to use dynamic range compression (see: Audacity Manual)

This may be a newbie question but is there a “perfect” Gain setting for these microphones or is it all based on what you are doing and how loud you will be?

I know the natural reaction is to put a piece of tape on each knob and that’s exactly where they’re supposed to be, but it doesn’t work like that with a live performance.

You have to set things for your exact speaking style, spacing, volume and room, and then you can’t change any of that during the show. You have to not get so loud that it overloads and starts getting crunchy, or so low that you start speaking the same volume as the natural microphone noise, buzz or hiss.

Play this short sound clip.

You can almost ignore that funny noise in the background, but if my voice got quieter, you would have to boost the volume and the noise would get louder, too. That’s a noisy microphone as an example.

All this comes with the low-cost USB microphone. That’s not how the grownups do it. They were shooting a video at the club last night. The sound guy set up each microphone and balanced the voice volumes for each shot and then checked volume and quality during the performance on his portable sound mixer. I missed getting a picture. Shucks.

I’m not sure how he was getting any dialog at all past the loud music, but this is me being glad I’m not him.

So if you’re doing an actual theatrical performance, maybe a simple Yeti isn’t going to do it. Or, like I said, you have to “Hollywood it” and make it sound like you’re having a fight while not moving around at all.

All those “Old Time Radio” shows with full cast were done with one or at most three microphones, one for the sound effects guy. They got good at backing away from the microphone when they had to scream or yell.

That’s why a quiet, echo-free room is a big deal. If you back away and your “studio” turns into kitchen echoes, that’s not going to work.

I do it with moving blankets.

Once you get rolling and complete a mixed-down performance, post it or the link to it and we’ll rip it apart. Don’t let that scare you. A number of people have been through here and made successful shows when we got done. Oh, and remember, it’s a family forum.


I don’t mean to say that being your own sound mixer is impossible. It’s not. But you do have to pay attention. Make the Audacity sound meters bigger so you can see them out the corner of your eye while you’re performing.

You can detach them and make them wider, too.
(Ignore the bottom 2/3 of that. It’s an illustration for something else)

You’ll get the hang of changing your voice volume to make the meters happy and still pull off the show.

It’s rough to do two different graphic things on the same computer.


I’m catching these as I think about them.

Putting the headphones into the Yeti may be the only way to do production without delays in the sound (Latency).

I have no idea how to effectively use the compressor to be honest.

You have to hit the performance close before you apply corrections and filtering. Audacity doesn’t do anything in real time, so it doesn’t matter what kind of software you have, it’s not going to prevent you from overload by yelling.

Just to cover it, you should be speaking into the BLUE logo and the performance volume is the knob on the back not the one on the front. You should be using the heart or kidney-shaped pattern. Yes, having the microphone up and slightly to one side is a good idea, but don’t go overboard or you’ll start to louse up the crispness of the voice.

Listening to your performance in real time on the headphones is an excellent way to catch yourself when you’re trying to be too expressive and nutso. That takes getting used to, too.

Did you go to any of those YouTube videos about how to make a podcast before you started this? They’re not all college-level courses, they almost all have good ideas.


Okay let’s see if I have grasped all of this information…lol

The Blue Yeti doesn’t have anything that pops up when it’s getting too loud. Just a red light on the front to show whether the mic is muted or not…it’d be real nice to have an indicator to show me my levels.

I’m also currently only using one monitor so when I’m full screen in a game I can’t see anything…unfortunately. I filmed today and attempted to back away and get closer to the microphone but some of my issues were still there…which I’ll get into.

I can always hear when I open my mouth or breathe…it only goes away with the gain all the way down. I have asthma so my breathing is usually a little wheezy but I try to keep that under control most of the time. I just want to be able to have that not happen with higher gain qualities but I’m sure that will involve controlling my breathing more than editing.

Since my room is so much larger than a normal “studio” do you think putting some kind of curtain behind me will help? I just want it to not be so loud but I currently don’t have the money to drop 300 bucks on more sound proofing foam. So maybe even building my own little box for the microphone? Also, if you feel like checking it out, you can see the size of my room here: Thats the size of my room and that’s all I have for foam right now…which is why I’m curious if putting a curtain right behind me will help or not?

And as for the staging arguments…it’s not that I’m trying to act out something. It’s me sitting in a computer chair yelling at a video game…so whatever I am screaming about, it’s generally because something frustrates me in the game I’m playing (Hence the full screen not being able to monitor audacity because I play games for youtube with only 1 monitor…equals problems)

I’m sure that the Dynamic Compression is going to really help out with my audio I’ll just have to read up a lot to make sure I know what I’m doing

it’d be real nice to have an indicator to show me my levels.

It’s not just nice. It’s required.

when I’m full screen in a game I can’t see anything

Then you’re dead. There has to be a volume indicator somewhere. You are a candidate for recording dialog on Something Else and not try to make one computer do everything. Two computers or the game computer and a stand-alone recorder like the Zoom H series. If you have a personal recorder on your cellphone or iPod, that’s another possibility.

I have asthma so my breathing is usually a little wheezy

You are rapidly talking yourself out of doing this. You can try Effect > Noise Gate to duck the spaces between words, but it’s rough to make that sound natural. Is Noise Gate a part of the regular Audacity? I’m never sure which effects are natural. I’ll find it.


I thought you said your microphone was up and to one side? That’s not awful, by the way. Everybody thinks you need to kill all the walls in a room, but you really only have to kill opposing walls, and the most important one is the one directly behind you, which is where the microphone is pointed. The microphone sees your face and everything behind you. Another soundproofing item is effects sound. If you have music or guns or explosions in the show, that goes a long way to covering up room echoes — although you still have to do very well. You still can’t record in the kitchen.

Remember when I said to record with the heart-shaped pattern? That means the place directly behind the microphone doesn’t pick up anything. That’s where you put the computer.

something frustrates me in the game I’m playing

At the risk of pounding on this, you can’t actually get frustrated. You have to act like you’re frustrated and control your volume and breathing. You start actually screaming into the microphone and the sound is going to turn to trash.

There’s another problem that’s not going to sink in until later. How many hours is it going to take to patch up your recording? A fuzzy rule of thumb is that production takes around three times the length of the show. So an hour game will take three hours to produce, edit, compress and export, four hours total. That’s if everything goes OK.

You need to figure out a way to set levels during the show. The other problems are just moving stuff around so it all works. Setting levels is killer.

Koz (4.13 KB)

Oh, and take the sock off the microphone. Listen for pops or thumps when you say “P” words. If you have the microphone up-left, then there won’t be any.

This is doing it wrong.


Okay I was mis reading the “acting” thing, I get what you mean now. I took the sock off and it still sounds the same no popping…which is good…yes? Here is a picture of my set up and how I usually have my microphone

Is that decent? Any changes? And here is the latest recording I did while playing a game, I think it came out A LOT better than my others, I moved the mic so when my “normal” loudness came in it hit at the .5 area and I tried to make sure to back away when I yelled…tell me if this sounds any better…this is with the compressor on it

Please tell me I’m making progress lol…I’ve been trying to make sure things sound a lot better and I’ve been controlling my breathing…basically making sure my nose doesnt whistle, I dont wheeze or anything. I have background game music playing while I talk as well, if you want to see and hear that it’s this one:

And thank you for all of your help so far…I will master this someday, I swear!

That seems to work. Good clear voice. No odd room effects, no echoes, good presence. No clipping or overload. I can’t tell where the game sound is supposed to be but “sound under” works for me (that’s the production name for the YouTube clip sound balance).

I think we’re here. I don’t remember what we were fixing, but there’s nothing wrong with that. I like the background soft orchestration of the game. I turned the volume up a little just to hear what was there. That’s perfect to establish the mood of the game without hitting you over the head with it. It also covers up what tiny microphone problems you may have left. I didn’t hear any.

I do have one “theater” complaint. You commented several times on camera about the loud game sound, but the real-time audience doesn’t hear that. That’s “Continuity.” Everything has to hang together. It’s oddly easy to screw up. If you produce a game over several days and cut it together into one show, make sure nobody says: “Wait, wasn’t he wearing a blue shirt a second ago? Why is it red now?”


Oh, one more. I wouldn’t use my potty mouth in public presentations.