Say I wanted to do some amateur vocal recording at home and try to get as close to studio-quality as possible. I've looked into it but it all seems quite complicated. Basically, what kind of processing equipment do I put in-between my microphone and my sound card to have it not sound like I'm talking through a wall?
If you're going to post something like 'first get a decent microphone', don't, then go away.
>>24763 >If you're going to post something like 'first get a decent microphone', don't, then go away.
If you don't want sensible advice, then go ask in /iq/ instead.
You chuck your soundboard out of the window, and you get a proper USB/Firewire interface. Something like an M-Box or Creative E-MU depending on your budget and software preference.
Then you get a semi decent condenser microphone. I'll let you research that yourself, but to give you a rough idea of range, I have an AKG one that cost me about £250 and it sounds completely professional, spending more than that is just getting towards silly elitist more money than sense territory.
Lastly you're going to want a decent recording environment. Depends on what exactly you are recording for, but usually for vocals you want to deaden the room as much as possible. Look up how to set gains properly and optimal distance from the microphone..If you can afford it, get a small vocal booth to wrap around the microphone holder, and a pop shield. Then hang duvets all over the walls of the room you are using.
The thing about making good sounding recordings is that it's more than the sum of the parts. Your gear doesn't have to be expensive as long as you do things properly and have the kind of ingenuity to work out a solution to the issues you might encounter- Some people can have all the expensive gear in the world and still come out with recordings worse than the EP I did as a teenager in my bedroom.
It puts them into your computer, to put it simply, properly. The standard PC soundboard is naff at this and while it technically doesn't affect the quality too much, it will affect the sampling and overall dynamics. Interfaces also typically have much better pre-amp for the mic- Basically that ties into what I said about getting the gain levels right and getting yourself at the right distance from the mic. These factors all need to be in balance, otherwise you will range from a boomy cave sound to a muffled underwater sound whilst still achieving technically the same quality.
I remember when I first forayed into recording. They told me I needed an interface, and I said nahhh I don't. Then after several nights struggling to get anything useful out of my cheap Shure mic with 3.5mm adapter and SoundBlaster Pro, I reluctantly went back to the shop and bought an interface and an SM57.
>>24767 A USB interface converts an analogue signal (from a microphone) to a digital one. Your computer will have all kinds of EM interference that will add noise to an analogue signal if it is fed in directly, but external interfaces allow proper shielding.
That's because it is. Audio recording is an art in itself, if you want professional results then you need to invest the time to learn. Without a reasonable understanding of the theory, you won't be able to get consistently good results.
>What does a USB interface actually do to the audio signals, though?
The same thing as your soundcard, only at higher quality and with the electronics needed to operate with proper microphones. Studio microphones are very different to the electret capsules found on gaming headsets and require careful preamplification. The type most generally suitable for vocal recording (condenser microphones) need 48v phantom power, which isn't supplied by consumer sound cards. The better electronics in a dedicated audio interface provide higher dynamic range and lower noise.
If you only need to record one thing at a time, there's a simple and relatively inexpensive option - a USB microphone. These are studio-quality microphones which have a preamp and audio interface built in.
There are a variety of these mics with a wide range of prices. At the lower end, I'd suggest the Samson Meteorite (£20) or the Samson CO1U (£55). If you're looking for bona-fide studio quality, then you'll need to spend a bit more - about £120 for an Audio Technica AT2020USB, a Rode NT-USB or an sE X1.
As >>24766 says, once you've got a quality mic then you need to sort out your room acoustics. Hanging a duvet behind your head will get you half way there. A Reflexion Filter (£50 - £120) will do wonders if you have the budget for one, or use another duvet behind the mic. A pop filter is essential with a condenser mic, but you can make one yourself with a wire coat hanger and an old pair of tights. You'll want to locate your computer as far away from the mic as possible to cut down on fan and hard drive noise.
If you want to learn more, consult the archives of Sound on Sound magazine. It's the definitive publication on home recording and their articles cover pretty much everything you need to know about recording, editing and mixing. If you'd prefer to have the information laid out for you in a course format, then I'd suggest the books Recording Secrets for the Small Studio and Mixing Secrets for the Small Studio by Mike Senior, which are the best introductory texts available.
Very cheap soundcards don't have balanced inputs or phantom power, which is necessary to use a microphone of any real quality. For serious recording, the cheapest useable interfaces are the Beringher U-Phoria series, starting at about £35. For that price you get a useful selection of software. A good value option is the Focusrite Scarlett Studio bundle, which includes an audio interface, a condenser mic, a pair of headphones (all of very good quality) and some software for about £140.
Speaking of software, there are now some surprisingly good free options, namely Mackie Tracktion and Presonus Studio One.
>>24770 >cargo-cult
Well, you can hear for yourself by plugging in a mic via the analogue 3.5mm jack. Although as >>24771 alludes to, impedance matching is also an important factor.
>>24763 You could get away with a decent soundproofed room using eggboxes. Go get a dictaphone, they have a decent enough microphone to be able to get an acoustic version of whatever vocal it is you're recording. And then stick on a banging donk on it and you're sorted.
>>24790 So, basically, it is just cargo cult nonsense? There are good reasons to use an external sound interface, but "no interference" isn't among them.
No. Computers are electrically very noisy so internal interfaces are susceptible to interference, but it's fairly low down the list of "reasons why consumer-grade soundcards are inadequate for serious recording use".
High quality PCI audio interfaces were relatively common before the advent of USB 2.0 and Firewire, but they have now died out almost completely. These interfaces commonly used external breakout boxes for the analogue circuitry, which reduced interference and provided more space for connectors. The only contemporary example of an internal audio interface I can think of is the RME HDSP series, which are digital interfaces designed for high bandwidth connections to digital mixing consoles via MADI or ADAT.
I think both are right in truth. Out of all of the many reasons why the mic input on an internal interface isn't suitable, interference is probably the least significant.
>>24793 A mate of mine does some VO work from home and all he does is climb under his duvet on the bed with his mic and a home made pop filter. Just use a wire coathanger bent into a square and stretch some tights over the wire square then zip tie the tights so it's taut over the frame.
>>24822 Although laptop audio can be noisy as all hell - far more compromised than a $5 PCI sound card, which implies some pretty impressive not-giving-a-fuck at design time.
I've purchased a CO1U Pro on the recommendation of >>24771, and am recording with a duvet over my head on the recommendation of >>24827. Being a single bloke, I don't have access to tights to make a pop filter with yet but I'm sure I can scrounge some.
I was frustrated to find the Samson sounded just as muddy and shite in Audacity as the dynamic Sennhauser plugged into the 3.5mm jack I was using before. Then I noticed people on the web talking about recording at a 48000hz sampling rate, and mine was set to 16000hz. Changing it to the former seems to have done wonders to the clarity.
Are there any more amateur mistakes I might be making in the recording software's settings?
>>24860 >Then I noticed people on the web talking about recording at a 48000hz sampling rate, and mine was set to 16000hz. Changing it to the former seems to have done wonders to the clarity.
Holy shit lad. Please tell me you understand why this change did what it did.
And after that, up your recording quality to at least 96000hz.
>>24861 There's not really any need for him to go above 48kHz, unless he's going to be pitch shifting or otherwise messing with the voice afterwards, but yeah, put it up anyway if the card supports it. (Nyquist frequency at sample rate of 48kHz is 24kHz, i.e. perfect sampling of all audio up to 24kHz - which is already above the range of human hearing.)
>>24861 >>24863 Won't upping the sampling to that needlessly large level increase the file size of my raw audio? I'd rather not do that if I can help it.
>>24862 Now that you mention it, I'm not, I am (was) using the motherboard jacks. I can't remember why I don't use the sound card I got off eBay with all its 5.1 bells and whistles but it's either because it broke soon after I started or the drivers were a hassle to use.
>Are there any more amateur mistakes I might be making in the recording software's settings?
As you've discovered, the sample rate determines the highest frequency that can be captured by the converter - set it too low and you'll lose the high frequencies, resulting in a muddy sound.
The other half of the digital audio equation is bit depth, which determines the dynamic range - the difference between the loudest and quietest sounds that can be recorded. The Samson C01U supports 16 bit recording, which is perfectly sufficient if you set your gain correctly.
Gain is the amount of amplification applied to the signal. Set the gain too low and you'll get too little signal and too much noise, resulting in a recording that is thin and hissy. Set it too high and the converter will run out of bit depth, resulting in horrible distortion.
You can set the gain correctly by looking at the input meter on your recording software. You want the peaks to reach -10 to -15dBFS (often indicated in amber) without ever hitting 0dBFS (usually indicated with a red "clip" light). You can check that you've set the gain correctly by looking at the waveform view of your recorded signal. It should fill a good amount of the waveform bar, without the peaks reaching the edges of the bar.
Be aware of ambient noise. Try to record in the quietest environment possible and put some distance between the mic and anything that makes noise, like the fan on your laptop. Getting closer to the mic will increase the amount of signal that it picks up, reducing the effect of background noise. For vocals, a distance of 10-15cm is usually best. Remember that the microphone is directional - it picks up sound most strongly from the front, and only weakly from the back. This directionality can be used to your advantage.
A proper stand is invaluable in getting microphone placement correct, especially if you're recording instruments like acoustic guitar. A fairly small difference in microphone placement can make a huge difference for instrument recordings.
It's worth watching or reading some tutorials for your audio software of choice. IIRC the Samson C01U comes bundled with Sonar LE, which is a perfectly good piece of software. You can find the documentation at the link below.
You don't need to record at a rate greater than 48kHz. Higher sample rates can occasionally be useful if you're doing extreme processing, but the idea that they improve sound quality in any meaningful way is an audiophile myth.
>>24866 I think it's better to not need it and have it than to need it and not have it. As you said, probably not necessary but it wouldn't hurt if you can do it.
In 50 years time when these recordings are released as original bootlegs from greatest artist the world has ever seen, they will be thankful of that extra quality.
The human ear simply cannot perceive the additional frequencies that are captured by higher bit rate recordings. You gain nothing, but lose disk space and CPU resources. High sample rate recording is only useful in weird fringe cases like extreme timestretching, but even there the benefits are marginal. High sample rates can actually sound worse in many cases, due to intermodulation distortion caused by the ultrasonic components.
24 bit recording is very useful, but only because it makes gain management easier. It's a convenience for the engineer that doesn't improve sound quality.
>>24868 Yeah exactly. This isn't an mp3 vs flac argument where there could technically be a difference in what you hear (despite double-blind tests showing this isn't detectable at the highest mp3 bit-rates), it's a case that your ears can probably barely hear up to 18kHz let alone above 24kHz.
The first sound file, the steady tone at 17.4kHz, I can hear in my right ear, but I can't hear it in my left ear. This scares me. Does it mean I've damaged my left ear somehow? Or could it just be my sinusitis throwing me off?
Everyone's hearing is a bit damaged, that's the point of the mosquito device. Ageing and noise exposure gradually reduce your high-frequency sensitivity, working from the top down. Teenagers can hear frequencies that older people can't, it's just a fact of life.
Your left ear might just be gunked up with fluid or wax, it might be ageing slightly more rapidly, or you might have suffered a blow to the head or acoustic damage. Drummers develop hearing loss asymmetrically, because the left side of their kit is considerably louder than the right.
The very high frequency part of your hearing isn't very important, so I wouldn't worry about it too much. A typical cassette player rolls off above 12kHz and a telephone doesn't respond above about 3.5kHz, so you can suffer a substantial amount of high-frequency hearing loss without experiencing significant functional impairment. If you're really concerned, go to an audiologist.
>>24870 At first I thought it was a stereo problem, but if I wear my headphones the wrong way around I still get the same result. Same as you, I can hear the sound in my right ear but not the left.
I used to be a drummer, though, and spent a good chunk of time with a headphone over my left ear as a monitor in other situations, so if above commenter is correct then that's a pretty plausible explanation.
Think about the layout of your kit. On the left side, you have your hi-hats and snare - they're very loud, and used almost constantly. On the right, you've got your toms, which are much quieter and generally only used for fills. The ride is less commonly used than the hats and generally played fairly quietly, as are other cymbals you might keep on the right like a splash or china.
>>24873 It made intuitive sense, but since I know nothing about hearing loss I didn't want to make my post sound like a confident "that's it". Thank you for the additional deatail, though.