Looking for audio recording advice
Some background here - I have a niche YouTube channel, mostly doing longform (often 3+ hour) commentaries on a board game with a friend. We're both very skilled at the game and are apparently entertaining to listen to, but it's all very amateur - I don't really know anything about videomaking except what I've picked up as I go.
That had never really been an issue, because the topic is so niche that only people really interested in the game watch, and they didn't really mind. But last year we got a shout-out from a big YouTuber and the subscriber count has gone up a lot since, to the point where it feels like I really should figure out how to make it a little less amateurish.
I want to start with audio, because that's 90% of what we do. But I don't even know where to start with it.
I got Blue Yetis for both of us a few years ago because I was vaguely aware that was a good mic, but if anything it made it slightly worse than just using headsets (quite a bit more echoey) - I don't know if that was a mistake, and if there are better mics to go for?
I'm also aware that you're supposed to have some kind of padding on the walls to stop the echoing, but that isn't really feasible for me both because I rent, and because all the rooms of the apartment I'm in are absolutely massive. I've heard that you can throw a blanket over your head, which I've tried for a short video but I absolutely couldn't handle for a three hour one. I'm lost on how to solve that too.
If anyone here knows much about it, I'd love either direct advice about it or a recommendation for how/where to learn about this stuff. It's super appreciated!
Because you rent this might be unfeasible too, but if you have a way to hang just a few things on some open/blank walls, or find a way to hang them on something portable (clothes rack or similar), large moving blankets are a good inexpensive way to get at least some basic audio treatment and improve the sound of the room. They're cheap and cover a large area. They're not as good as actual treatment but surprisingly good (I use them in my home studio).
Basically any possible hack you can think of to safely get a lot more soft (pillow/blanket/clothes/fabric/etc) surfaces in the area the better to reduce the amount of sound that is going to bounce off hard surfaces and cause the echoes
The thing about the (basic/gaming) headset mics (in my anecdotal experience) is they're extremely directional so they're picking up a lot less of the room's sound whereas the Yetis, while a decent basic upgrade as far as mic quality, are going to pick up much more of your surroundings.Apparently this is incorrect and more just an effect of proximity. See replies.Tiny mics in headsets are almost impossible to make very directional outside of high frequencies, generally the methods to build directional mics need either bigger capsules or multimic arrays (which can be quite tiny with MEMS microphones, but that's not what cheap gaming headsets normally use). The effect OP is talking about is almost always just the effect of proximity: when the mic is just a few centimeters from the mouth, voice is so much louder than the room that it effectively suppresses noise.
The Yeti is going to be similar when used from such a close distance.
Gotcha. Edited my original post to reflect this. I am certainly no expert.
Not sure so much with gaming headsets, but I do know professional headset mics are frequently omnidirectional pick up patterns. So the reason they pick up less room noise is not so much the pickup pattern, and more so the mic placement gets you a better signal to noise ratio.
This is a fantastic recommendation. I have heard of gaming creators cover their desks in pillows to help reduce the echoes.
Ah yeah I was particularly talking about just basic gaming headsets, most that I have seen and have owned have been very directional (sometimes too much, to the point of having to mess with it a bunch to get a proper pickup from it). That said it has been years since I have had any because I always use a proper mic setup now whether for voice comms or for recording
I do have a condenser mic in a relatively untreated room but I keep the gain reasonable and the mic just a few inches from my face and that combo seems to sound quite good without picking up too much of the room
My studio has all walls covered in blankets and band flags and a carpet floor and it sounds even better and surprisingly good considering the makeshift nature of it
I do have clothes racks, I can try this - thank you!
Quick question on that front - I assume they don't need to be against the walls, they just need to enclose the recording area (which would be a lot easier since the room is very large)?
Does there also need to be something between me and the ceiling?
Hi! Fellow small youtube creator here.
The way you can picture sound is like bubbles. Something makes a noise, like your mouth, and then an air bubble expands into all directions from that point. All hard surfaces like walls, desks, computers, kitchen fronts, etc, will echo this sound back very well, which is what's picked up by the microphones and makes your voice sound echo-y and hollow.
So yeah, sound will reflect from the ceilings as well. There are a few things you can do:
I don't know enough science behind it honestly to tell you for sure the effectiveness difference between those, but even if they're surrounding and not up against the walls it should still help because it's getting in-between the source of the sound and the wall. Given a very large room maybe it's better have it all as close as you can get it so there's more chance for sound to hit it first before hitting anything else hard in between- basically reducing the overall untreated surface area available for the sound to hit
People do sound treat their ceilings- I have gotten away with not doing so for my own purposes. Probably very much depends on the space (which is true for all treatments!) but often it's a bit more difficult to deal with ceilings on a budget (at least i never thought of good ways to achieve it myself but also didn't bother too much)
I do know that it is effective for there to be an air gap between the soft surface and the hard surface. I was doing live sound at a venue once that made their own acoustic panels with fabric, 2x4s, and rockwool insulation. Each panel had the depth to fit two sheets of insulation in them. However, when they were testing during assembly, they found one sheet worked better, and to leave a gap between the wall and insulation.
Make sure you’re speaking into the front side of your Blue Yeti, not the top of it. Keep your mouth just a couple inches away from it when you speak. Consider adding a windscreen to the mic if you don’t have one, that can help with popping sounds from plosives, and some wet mouthsounds. Not sure if you’re currently doing any post-processing but lots of A/V software these days has noise reduction, low/high pass filters, normalization, etc. These can all be helpful!
My desk is a bit low, so it's possible I was talking into the top of it. I'll stack the mic on some books or something. I'm not sure I can only be a few inches away though - that would be a bit close to the monitor for my liking. Does being further away tend to be a problem?
The reason to be closer is so that you can turn down the gain on the microphone, which will result in picking up less noise that is not coming from your mouth (e.g. echoes off the walls). If you're a foot or two away, in order to pick up your speech clearly, you'll have to turn up the gain enough that it is likely to also pick up a lot of sound reflecting off of the walls.
In my sound engineering course in college, our instructor talked about the 1:3 ratio: the primary audio source should be no more than 1/3 the distance to the recording equipment of any background audio sources (sorry that was a clunky sentence). Put another way, if you're 1 foot from your mic, your mic will also be clearly picking up sounds from your walls up to three feet away. If you're 4 inches from your mic, you can likely get away with other sounds that are more than a foot away.
There are some dubious claims around the thread. My specialization is sound reproduction, not recording, but I dabble in recording a bit as well, I have recorded and mixed a few songs for my band with good results.
I agree with most of your post, there is a bunch advice in the post that will make the OP spend money on things that aren't strictly necessary to level up their recording game using their existing hardware. Spending money on better recording hardware is an option, but it is more important to get the basics right in both recording habits and environment.
Some nice to haves that can help for OP though are:
Both aren't strictly necessary of course. Placing the microphone on a higher surface that isn't the same desk OP might be leaning on, typing on, etc will achieve much the same.
There is one piece of advice I don't entirely agree with.
This very much depends on the room layout and where OP is sitting. If there are uncovered walls nearby and even more so when they are sitting in a corner the reverb from reflection can still be noticeable. Having said that, it rarely is needed to cover everything with blankets. But having some padding can make a noticeable difference.
Considering OP mentions they have clothes racks this is fairly easy to experiment with. And specifically something I'd encourage as it is one of those things they can do without additional costs.
Thanks, I agree with all this.
The video you sent made me realise I have the Yeti set to the wrong setting - the one that looks like a figure-8 instead of the dented circle. So that's one change I'm going to make!
Also going to try to see if I can reasonably get closer to the mic without being too close to the screen. I guess I need to look into mic arms for this.
What is EQ exactly, sorry? I know embarrassingly little about all of this!
EQ is short for equalization/equalizer, which is a process, nowadays done with software, that changes the loudness of various frequencies. Generally microphones (and also loudspeakers or headphones) are never perfectly linear, meaning they record each frequency with a slightly different loudness and change the color of the sound in that way. They also often do not record the very lowest frequencies, some do not record treble very well etc. With EQ we can change the balance of various frequency bands and make the device more linear - if it seems to have a bit too much sibilance for example (as the Yeti might have), you can turn the frequency band where sibilance is located down a little bit and make the sound more natural. This can help the overall sound a ton.
However, a specialized EQ software may not be necessary here, as firstly part of what the person in the video is doing is tuning a very basic EQ that is built into the Logitech software, and secondly the overall color of the sound changes based on how close your mouth is to the microphone (this is called simply "proximity effect"), and moving closer generally increases the low frequencies, so the sound often no longer sounds thin or sharp. In youtube videos the Yeti seems to have okay sound when used properly and from up close.
This should help, that is the cardioid pattern and it picks up sound from the sides and back less than from the front. But the main takeaway from that is that I recommend reading the manual, because unless Logitech is really lazy it should explain these things and possibly provide some best practices on how to use it as well.
Get a few of those long bars used for photo backdrops since they come with heavy duty clips, you can use heavy blankets or acoustic blankets/sound proof blankets. When it comes to mics, I'll just say I haven't done some studio audio engineering in like 20 years but the standard was the RE20. It was used by singers, broadcasters and radio. It's the mic that you would see on The Howard Stern Show.
You probably want dynamic microphones. Because of how they work, they are less sensitive and have a more specific pickup pattern, which means rhat they don’t pick up background noise. The con is that they need to be right next to your face - like, literally less than inch away from your mouth. They’re used for live performances often, so if you see artists slobbering over their mic, that’s why.
Alternatively, you can try to get condenser mics with tighter pickup patterns. There’s also hardware options like noise gates.
If you have the space, I’d just get some dynamic mics and put them right up in your face.
Barring that, if space is an issue, get some lav mics (dji set, for instance) that are attached to the body.
I would argue that dynamic mics do not have tighter pickup patterns, it is just that they have a quicker fall off of volume as you get further away, The sides can be just as sensitive as a condenser, it is just that as you move away from a dynamic mic, the volume drops off quicker. I do agree though that in an untreated space a dynamic mic would probably be an improvement.
Noise gates can also be done in software. However, I would argue against using a noise gate. The reason is for example if you have an HVAC hum, after a few minutes your brain will tune out the hum. However, with a noise gate, the hum would be suddenly cutting in and out when the person talks, which makes it more jarring and something the brain doesn't tune out (and might even notice more than doing nothing with it at all). If needed, the better option would be to use an expander as it is similar to a noise gate, just with more subtlety.
This is not true. And the quicker falloff is not true either afaik, they just don't really pick up the highest frequencies, so with the (normal) falloff with distance the already muted high frequencies vanish early. But the way I understand it, the only reason why they tend to produce less feedback or room sound is that they're usually used from up close and therefore with less gain, and the reason why they're so popular for live sound is that not having the highest treble doesn't matter as much and they're much more durable than condensers, plus they don't suffer from handling noise as much.
At home where feedback is not an issue a condenser with a pop filter used from up close tends to sound better for less money.
In addition of hardware fixes, you could "get" a copy of Izotope RX 10 or similar and use the de-reverb feature in there, it seems to work quite well.
Also check diy booths : https://www.youtube.com/watch?v=rWgLCPaOAzo
The neural net models in UVR, which is a free piecee of software, are pretty great as well. But all de-reverb software does sometimes produce artifacts. Sometimes adding a little bit of artificial reverb back using some free VST plugin is enough to mask the artifacts.
As others have suggested, you can hang some heavy blankets up and it'll go a long way.
You may also want to look at some post processing. I'm not sure what you're using to edit your videos, but I think most current programs can use VST plugins. I do music and not voiceover work, but I'd recommend De-Verb (pluginalliance.com) if there's still too much reverb after the blankets. You may also want a tool for easy vocal processing. It's been a bit of a trend lately for every big company to release their own vocal all-in-one plugin, but the only one I've used is Universal Audio's Topline Vocal Suite which may be more than you need. Another good option here is Novel tech's Vocal Enhancer, although it doesn't give you much control. It's a one knob sort of thing that cleans up vocal tracks and is also on Plugin Alliance. (BTW, don't buy anything on Plugin Alliance at full price. They're always running sales.) I would look into threads about podcast mixing. I'd imagine it's all the same principals for videos.
So I use auphonic.com to clean up my audio for all my youtube videos, I'm amazed everytime I use it cause i've sent some pretty low quality recordings into it and it comes out sounding so good. A different approach then others have suggested but I'd check it it out if you can't control your environment very well.
The site doesn't look very fancy but it does well.
I spent my early life working on and off as a recording/sound engineer and playing in bands, so this is an area where I have some relevant knowledge.
Let's talk microphones first: You mentioned the blue yeti mics, which I've never worked with, but I googled and it looks like they're USB condenser mics. At the high-end, in a well-controlled environment, condenser mics are great, but at the lower/intro-tier and noisy/uncontrolled environment, I believe your money would be better spent on a pair of good dynamic mics. Dynamic microphones like the Sure SM58 (which can be had new for <$100 each or used for much less) are the world standard for live/ad-hoc vocal performances because they work amazingly well in a wide variety of environments, they don't need phantom power, and they're durable as hell (compared to other microphones).
Room treatment? You mention some unwanted echoing that only became apparent when you swapped from headsets to the yeti mics. I question if that's from the room or something related to the software packaged with the yeti microphones. I mentioned never using those mics (or any pro-sumer USB mics honestly) but I imagine they all come with some drivers and a software package for recording. I suspect that they come with a default profile of effects, equalizer settings, impulse response, etc. What I suspect you're noticing is some in-baked reverb in those settings, but I can't say for sure. What I can say is that you want direct control over the post-processing (more on that later), and also that room treatment can get insanely expensive, is easy to do wrong, and rarely provides novices the results they want. There are lots of good videos on studio design and room treatment on youtube. Since you're just recording vocals at a speaking-volume, the "avoid hard parallel surfaces" concept is sufficient. I would suggest setting up everything else first, see what you get, and maybe look into some rugs or hanging some stuff on the walls if your room geometry is really the source of unwanted reverb.
After the mic, or: How I Learned to Stop Worrying and Love Reaper. Right now your interface is built into the mic. Aside from swapping mics, the other thing I'd really suggest for someone in your shoes is picking up a cheap two channel audio interface (analog to digital converter) like the focusrite scarlett. This gives you the ability to use "real" mics, gives you much more control over the preamps, and leads you to recording using a proper digital audio workstation (abbreviated DAW). The biggest change for someone wanting to step up in quality is gonna be learning a DAW. Some cost a lot of money up front, some cost a lot on a subscription, and my favorite one (Reaper) is open source and free. They are all complex pieces of software with significant learning curves. I'd suggest just throwing yourself into learning one of them. Reaper is free, it's more than "professional enough" for plenty of professional recording studios, and its workflow is great once you get comfy with the hotkeys. Over time you'll acquire a bunch of plugins, and if you "get good" at the software side of recording, then you can truly work magic. However, just getting something recorded and sounding good with Reaper is not hard (an hour or two of setup and learning), and you can grow from there.
I know that was a lot, so in summary: I'd get good dynamic mics (like the SM58), a cheap 2-channel audio interface (like the focusrite scarlett), and focus on learning a DAW (like Reaper). Room treatment can wait; learn to record with the DAW and watch videos or read about room treatments.
Other things to consider: cables, mic stands, pop filters, etc. Don't get the cheapest cables, but don't buy the gold-plated crap. $15 for a short XLR cable is reasonable. Unless you're moving around a lot, probably a simple stationary mic stand for each of you is fine, but handling the mic while recording is a big no-no cause "clunk clunk". Pop filters can be helpful, but you can make them out of pantyhose and should not spend big $$ on them. Patience while you learn the DAW is going to be the hardest thing to come by is my guess, so find a youtube guide you enjoy watching.
Possibly this is a nitpick, but it's neither. It's not open sourced, and it has a fully functional trial period of 60 days, but those 60 days are not enforced by the software, you just get a nag screen on start up, with which it is technically unlimited. But I think it's worth saying that it is still a commercial piece of software and you are asked to buy it after the trial period. The price makes it such a good deal (60 USD for a discounted version which applies here) that it's worth supporting it.
An alternative: we just switched to sE V7 with my band because it's a bit more flat, with less of a presence boost and a slightly more extended high end, which makes it sound slightly more studio like. It costs slightly less than an SM58. But honestly I think the Yeti might be okay, OP sounds like they're not using it exactly right, it should sound good when used from up close with the cardioid setting even with non-ideal acoustics.
One thing that can help reduce the echo when using a condenser mic like the Blue Yeti without going all out hanging stuff on the walls or sound-proofing your whole recording room is a sound shield. Basically a padded tri-fold kind of thing that you can set up on the opposite side of the mic from where you're speaking into it to catch and dampen your voice before it has a chance to go echo around your room. You can get mic stands or arms that have them built in (just search sound shield on Amazon to see the kind of things that are out there). They're definitely not perfect--they can be a bit bulky and cumbersome, and you'll likely still catch echo and ambient noise, but in my experience it reduces those enough to make it a lot easier to clean up in post processing and get very clean sounding vocals.
Some good nuggets of advice in the comments already. I'll just reinforce a few points. I co-hosted a podcast in the past (39 episodes), and do live audio regularly, and (amateur) recorded audio from time to time.
You can do some self study about what each of those is. If you do end up using Ardour, I can tell you exactly the plugin stack I used for my podcast.
I see a lot of people recommend different microphones. Before you go for that, make sure the basics are in order first. You can have the best hardware there is but if you don't use it right in a less than ideal environment it will not make that big of a difference.
Having said that, if you ever do decide to go for a different microphone, there is some additional information to be aware of.
What isn't mentioned too clearly in other comments, is that all these fancy microphones need additional hardware. Specifically an audio interface. Most of these microphones have an XLR connection and need an audio interface. Audio interfaces connect to your computer over USB and will show up as a microphone and soundcard. The soundcard bit is handy because they also have a monitor function. Which allows you to listen to yourself recording without a delay. This gives you a better idea of the quality while you are recording. So you just plug in your headphones in the audio interface, use it as a regular soundcard for playback and use the monitoring function during recording.
There are a lot of (cheap) options out there. One you will often see recommended is the Scarlett Solo, which is a pretty good audio interface. Though, the third gen had some issues with distortion on phantom power for some people (including me) not sure how that is with current (fourth gen). Because of that issue I switched to the UA Volt 1 which has been great. Though I am not a fan of how they lock their AISIO drivers (whole other rabbit hole) behind a portal with account activation.
One last "fun" fact about audio interfaces. Discord doesn't play nice with a lot of them. They often expose dual channel audio where only one has the microphone audio (the other is for instruments). Discord downmixes it to a mono signal but in such a way that the volume is extremely low. I had to resort to setting up voicemeeter to sit between discord and my interface to do the mono downmixing properly. Which is a bit much, so I figured I'd mention it if you also use Discord.