You’re probably here because you’re getting more serious about podcasting, recording music, or general audio engineering and want to take the next step in getting your final product to sound better than it does today.

Audio interfaces are confusing, not just when you’re starting out, but also when you have been into music production for a while.

Just like instruments themselves, audio interfaces come in all sorts of variations; purpose built for recording, live shows, general or hybrid use; and some have built-in effects, and others are as simple as a small box with one input and one output.

That’s why we’ve compiled this comprehensive guide on audio interface I/O for you to bring an end to not knowing what any input or output does.

Different configurations for different types of Audio Interfaces

There are many different types of audio interfaces that are a better fit for different people depending on their use-case; e.g., recording and mixing a live show, recording a podcast, or just being new to audio engineering in general.

One of the first major factors that distinguish audio interfaces from each other are the way that they connect to your computer. As a disclaimer, none bests the other necessarily, but at the same time, depending on use-case, one can definitely be the better option for you.

There are three major attributes that distinguish Audio Interfaces from each other: 

1. Connectivity Types
2. Types and Numbers of Inputs/Outputs
3. Knobs/Controls  

I’ll also get into some more advanced features on interfaces, but if you’re just starting off, these are certainly the three criterion upon which you would want to decide on what type of Audio interface to purchase.

Connectivity Types

USB

Classic USB connections are probably the most universal connection types, but are typically slower than the other connection types.

With that said, USB is stepping its game up with the rise of USB-C inputs on computers which allow for data to be transferred more quickly than your standard USB (in this case, the data is sound). 

A classical example for an USB audio interface is the Presonus Audiobox USB. Which is the same model this guide is illustrated with.

FireWire

While most computers don’t have a FireWire port, FireWires are typically faster and are found on nicer, higher-end audio interfaces.

Unless you are already computer shopping as well and can take this into account when purchasing a new laptop or desktop, you will likely need an adapter to support a FireWire audio interface, unless you go with a hybrid like the MOTU 4pre USB FireWire Audio Interface.

Thunderbolt

Sometimes compatible with USB-C, Thunderbolt is gaining popularity in becoming a standard port.

That said, the backwards compatibility isn’t always there, so be careful when choosing your computer as well as Audio Interface. Our recommendation is the Universal Audio Apollo Twin MKII Solo.

PCIe

PCIe connections are typically found in desktop computers as they are essentially a second sound-card that plugs directly into your motherboard.

Which one is right for me?

If you’re just starting off, any type of connectivity will suffice as long as it fits the budget.

That said, the most economic options are typically connected via USB.

Number and types of Inputs/Outputs: In short, the lower the number of I/O is, the smaller and more portable it will be and the converse is true as well: the more the I/O, the larger and more stationary it will be.

It is also fair to assume that the more I/O’s on an audio interface, the types and functions of these I/O’s will vary as well: some are designed for microphone cables, some are designed for instrument cables, some are designed for MIDI, and some are designed for speaker output, headphone output, etc. 

Inputs and Outputs

Standard Inputs

• Microphone: self-explanatory
• Direct Injection/Line Input: used for electric guitar, electric bass, etc.
• MIDI: used for midi keyboards or electronic drum sets Standard Outputs:  XLR (Balanced) – A balanced signal is almost always better.  What this means is that the XLR cable carries two copies of the signal with reverse polarity (not reversed phase). This is what is also used for microphone cables. TRS (Unbalanced) – One copy of the signal is sent to the DAW. This cable is also known as an instrument cable such as something you would plug in to your electric guitar. MIDI – sends the MIDI data to the DAW RCA – the red and white plugs you see on some interfaces or speakers.

Most audio interfaces have hybrid ports that support both microphone and instrument cables. You can see from the inputs on both my AudioBox and StudioLive mixer than all inputs have a hole in the center to plug in an instrument cable, while the larger shape of the port is designed to fit a microphone cable as well.

Standard Outputs

• XLR (Balanced) – A balanced signal is almost always better.  What this means is that the XLR cable carries two copies of the signal with reverse polarity (not reversed phase).  This is what is also used for microphone cables.
• TRS (Unbalanced) – One copy of the signal is sent to the DAW.  This cable is also known as an instrument cable such as something you would plug in to your electric guitar.
• MIDI – sends the MIDI data to the DAW 
• RCA – the red and white plugs you see on some interfaces or speakers

Most audio interfaces have hybrid ports that support both microphone and instrument cables. You can see from the inputs on both my AudioBox and StudioLive mixer than all inputs have a hole in the center to plug in an instrument cable, while the larger shape of the port is designed to fit a microphone cable as well.

Knobs / Controls

This one is hard, especially in my situation, where I have three or four different pieces of hardware that produce the ultimate “raw” signal to my DAW.

At your most basic level, you’re going to want to have these controls:

Input Gain

Also known as a preamp gain, you can adjust the level of the incoming signal.

These dials, just like the parameters on an amplifier, have a ‘0’ setting at 12 o’clock, and could dial all the way left and right to reduce and increase the signal respectively.

Adjusting this dial positively (+) helps for instruments that tend to have softer signals like passive pickups and some microphones.

These soft signals are what are generally referred to as low-impedance, or “low-z” signals. Impedance simply refers to the voltage of a signal.

Low-impedance signals have a lower voltage, lower energy, and therefore will need to be made-up for through adjusting input gain/preamps, though this can also be fixed through audio normalization (a software process in your DAW), or simply through applying gain post-recording.

Conversely, you would also likely want to lower your input gain (-) with high-impedance (“high-z”) inputs by comparison: high-impedance instruments that have active pickups will naturally come in louder since they have more voltage, or in other words, higher energy.

You want to be careful with preamp settings when you’re plugging in almost anything – especially if your speakers or headphones are typically on a louder setting at all times.

Having your preamp turned up too high while plugging in a microphone or instrument can cause a ton of loud pops that can cause damage since these sounds almost always exceed 0dB.

Audio Clipping Indicator

In general, setting preamps too loud will cause your audio signal to clip or allow too much noise to seep in and your track will be hard to remedy, no matter how much you process the signal.

If you’re having difficulty understanding when a signal is too loud, there is usually a red dot that will light up when the signal is too hot and is clipping or near-clipping. I know how tempting it is to be loud, but you must resist the impulse for now and try to record your tracks where the highest peak (i.e., the loudest single moment of a track) is roughly -6dB at most.

You can also look at the visual track of the recording itself.

Are some troughs and crests (peaks and valleys) getting cut off by literally extending beyond the frame of the box that the recording looks to be ‘in’?

The waveforms of a track will appear to be like a knife with the tip cut off at the top or bottom of any given track if so. Then you’re clipping – even if it’s only for a split second.

You’re going to want your track to be recorded more softly so that none of the signal is getting clipped, and you will make up for any desired increase in volume through effects processing gain staging, mastering, and so on. You usually can “normalize” audio to increase tracks appearance and volume to a good size, or you can always find a volume setting on the visual of the track again without lowering the fader.

Output Gain

Same rules apply for output as does input and these two controls should be used in tandem to sculpt a moderate signal.

Think about it this way: When you plug your phone into the aux port in a car, you want to put your phones volume on higher, but not too high that it sounds harsh. You also don’t want it to be too low where your car volume is maxed out but it just doesn’t feel quite loud enough.

You typically need to find a balance between input and output gain to achieve the signal you are looking for when recording.

Too much input gain can result in overly sensitive listening of a mic or instrument pickup, where you are boosting too much noise or the unwanted subtleties of an electric guitar.

Headphones

Just like when you play music in your car or on your phone – you don’t want to hear anything playing too loudly or too softly. 

Headphones will typically have their own controls separate from the output.

They essentially function in the same way without actually altering the signal itself, it just changes how loudly you perceive it – like a volume control on your smartphone when listening to music doesn’t change the loudness of the song, but rather just how loudly you want to hear it. 

48v – a.k.a “Phantom Power”

Phantom power is a function of most audio interfaces that supplies power/voltage to the signal source, such as a microphone.

This essentially means that a quality microphone can require power to operate, and will need to extract this power from the Audio Interface.

One thing to note is that, smaller interfaces extract their power from the computer and also have a phantom power button.

If you are using a laptop or some form of portable computer, it is wise to keep it plugged into the outlet at all times, but especially when attaching devices that do not pull from a power source of their own.

EQ

EQ, or equalization, is likely the most fundamental “effect” one can apply to any given bit of audio. However, if you apply this on a hardware level (rather than applying it through your software), it cannot be undone or altered without heavy compensation on a software level, and even then the track will likely sound unnatural, unless you re-record that same exact track.

So, while this can absolutely work to your benefit by not having to double-up on software effects and saving CPU/resources, it can just as likely work against you if you realize that the tone is simply unamendable to the mix.

For example, you may already have your bass frequency space being utilized by a kick, bass, and other instruments.

Recording a guitar with low frequencies boosted and cutting out high frequencies will not only muddy up the overall mix, but you will also be unable to salvage much of those higher frequencies with quality intact.

Recording with EQ applied (more than one knob or setting is not at ‘0’ or default), is referred to as committing to the effect due to the un-salvageable nature of a track recorded with an effect like EQ already applied.  Some of the benefits of this are being able to manipulate the existing frequencies, as well as utilizing less CPU resources through software plugins.

If you’re new to recording or mixing, you should definitely start off by not committing to any kind of effects or EQ boosts that are applied to the track as it records.  Instead, record the raw signals at default levels and no FX, then tweak it through your DAW using the stock EQ plugins and other effects that can be altered, removed, and re-applied.

EQ comes in all different formats which are typically dictated by however many ‘bands’ they have.  Frequency bands are essentially the sonic range in which the EQ knob can affect a signal.

For example, my mixer has a basic 3-band EQ for each channel.  The top knob has a predetermined frequency of 12 kHz that I can add 15dB to, or subtract 15 dB from.

The mid-range knob on my interface is particularly useful, as it allows me to set the middle frequency from a range of 140Hz to 3.5 kHz, and then raise or lower the signal from there.  Lastly, my low range EQ band yet again has a predetermined frequency of 80Hz in which I can yet again raise or lower by +/- 15dB.  

You may notice, as is in my case, that there are frequency gaps between each band: 80Hz, 140Hz to 3.5 kHz, and then 12 kHz.  What about everything in between?  Try to visualize pinching your shirt and pulling it out so that it forms a cone shape – unless your clothes are tethered, you aren’t pulling just one thread

In the same way that we aren’t affecting just one frequency.

While the boost is most concentrated on its frequency setting, it also raises the surrounding frequencies proportionally with less concentration.

SEND/FX

Sends exist on interfaces as well as software and are essentially auxiliary outputs compared to the “Main” out bus, but they function exactly the same. All sends and “Main” busses are essentially ‘parent’ busses for other busses and individual tracks.

For example: if you have all tracks (drums, instruments, vocals, effects, reverb, etc.) all going to the master bus, you will have to do a lot more work getting it to sound clear.  However, if you send all drum and bass to one ‘send’, all ‘band’ pieces to one ‘send’, etc., then you will find that these tracks all find their own space without interfering with other tracks as much and you can adjust their holistic levels separately.

You will still need to make edits though to procure that extra clarity and separation in a mix.

Some Practical Application

In short, you should always start with less gain on any signal, and then you can always add more until you’re peaking at -6dB give or take.

It’s hard to get this perfectly, and this is where effects like EQ, compression, and limiting can help taper those peaks without sacrificing the loudness or quality of a signal.

With the gain settings put all the way down, try plugging in a mic or instrument. Now turn up the gain settings until you are hearing the signal from your speakers or headphones.

If you have a powerful microphone, you will likely need to turn on the phantom power setting. Again, you want to plug in your cables and turn on this setting with the gain/volume controls set all the way down to 0 to avoid any damage being done to your ears.

Adjusting the output gain on your interface will also help you get there.

You may need to incoming signal to get boosted in order to register on the audio interface, but then need to lower the outgoing signal so that it isn’t clipping or too loud when it is recorded onto your software. The converse may also be true – you’ll need to lower the incoming signal and raise the outgoing signal as well as any combination in between. 

You also want to keep your speaker and headphone levels low.

This allows you to accurately perceive your mix as higher volumes will raise specific frequencies and signals disproportionately compared to others.

In addition, you could also fatigue your ears by listening too loud for too long, or even lead to hearing damage or loss.

What now?

Now that you are equipped with the

I really hope this article was helpful to you and provided you with everything you would want to know about audio interface I/O.

If this article was helpful to you, you would do me a great favor by sharing it on Social Media!