Transcript for:
Audio Recording and Editing Guide

Hello and welcome to this video walkthrough for project 2. The primary goal of this project is to familiarize you with best practices for using studio technology to record and edit audio. And as a secondary goal, this project also reinforces some of the signal flow and gain staging concepts from the previous project. To a lesser extent, this project attempts to emulate some of the practices and challenges that composers faced in the 1940s and 50s when their medium involved cutting, splicing, and manipulating reels of magnetic tape. With the advantage of computers and digital audio technology, your task will be considerably easier. Nevertheless, I hope this activity sheds some light on the difficulties of the early years of electroacoustic music.

Specifically, in this project, you will set up a microphone, record... The sentence you see here and then edit that recording so that it sounds like this sentence here. Step one, power on and log in. Just like in the last project, turn on the main studio power switch, turn on the computer and the display monitor and log in as composer. Don't forget the password, which is also composer with a capital C.

Check the Mac OS. system preferences and make sure the Scarlett is selected as both the input and output device. Step two, set up and connect a microphone.

Somewhere in the studio, you'll find a drawer containing two mics, two mic cables, two mic clips, and a pop filter. Take a look around. You should also find two microphone stands. If the microphone clips aren't in the drawer, and aren't in the microphone pouches, be sure to check if they are screwed onto the microphone stands because some sometimes people leave them there by accident.

Take one of the mic stands and set it up in a rough position you think is going to work well. You can adjust it later if you need to. You can record standing or sitting down, doesn't really matter, but I think standing might be a little bit better just so you don't accidentally capture some creaking chair sounds. If you've never done this before, you might have the urge to lower the central post so that it makes contact with the floor, but that's actually not the correct thing to do because three legs is optimally stable, and the central post doesn't have a rubber foot on it. So it's possible that it might transmit footsteps and other vibrations from the floor.

Before moving on, make sure all of the tightening posts on the stand are snug because you want it to be secure, but not... over tightened because you want to be able to easily adjust if necessary and trust me if you do over tighten becomes really difficult to loosen so once that's up take one of the mic clips and screw it on and then carefully clip one of the microphones into the clip and these are condenser mics so they are a little fragile so handle them with care I think the easiest and most gentle approach is to slide into the clip and first until it's nice and snug Depending on what studio you're in, the microphone might have an on-off switch, so do be aware of that. Now, let's talk pop filters. In a close mic studio situation, if you are unaccustomed to speaking into a microphone, it's easy to accidentally pop the microphone with plosives, which are speech sounds that involve a sudden burst of air. And against my better judgment, I'm going to demonstrate what...

popping the mic sounds like with some plosive sounds of my own like p b and k it's not a it's not a nice sound sorry so popping the mic produces a big obnoxious high amplitude low frequency thwacking sound and basically ruins that spot in your recording pop filters have the effect of diffusing a concentrated burst of air while remaining almost completely transparent from an acoustic perspective for the record if you don't have a pop filter the usual strategy is to position the mic so that it's slightly off axis of your natural airflow which is basically what you see here in addition you can also try to subconsciously train yourself to soften your plosives keeping the pronunciation clear while minimizing aggressive airflow Of course, you do have a pop filter in the studio, so a good spot for it is directly between you and the capsule, with about two to three inches on either side, making sure it doesn't touch the mic or you. Now, this isn't rocket science. The distances are flexible. You don't need to get out your ruler.

But this is typically what I do. You might find it kind of difficult to clip these particular pop filters onto the same mic stand that the microphone is connected to. The stems are kind of short, they don't really hold their shape all that well, so what I recommend to avoid frustration is setting up the other mic stand in a position that allows you to clip the pop filter onto that second mic stand and then position it so that the filter is in front of the mic.

Finally, take one of the XLR cables and just like on the patch bay, the correct technique here is to connect to the source first then destination. So take the female end of the cable and plug it in to the end of the microphone. You should not have to push very hard, and it should make a small satisfying click when you click it in.

If it doesn't click but it feels snug, it's almost certainly plugged in. So don't force it. And not all XLR connectors are going to have the same shape, so not all of them have this locking clicking mechanism.

Now we could take the other end and plug it. into the Focusrite audio interface, which is certainly what you do in a simplified bedroom studio setup or something like that. But since you're here, we're going to use this opportunity to reinforce signal flow concepts. So take a look around and find a panel with two female XLR connectors on it and plug the other end of the mic cable into one of these. The other ends of these connectors are wired directly into mic inputs 11 and 12 on the mixer.

So if you plug into the left connector, your signal shows up on input 11, and the other one shows up on 12. Okay, step three, route and adjust the mic signal. So first, take a moment to make sure the input channel strip on the mixer is in a completely neutral state, with aux sends completely down, and EQ completely flat. You're using a condenser mic, which requires phantom power in order to function.

The mixer... is already configured to provide phantom power on all mic inputs, which you can verify by making sure the little red LED labeled 48V is illuminated. The first gain stage in the signal chain is the gain knob at the top of the input channel strip on the mixer.

This is a mic-level signal, so it's going to need a considerable amount of pre-amplification to bring it up to a nominal line level. So, you know, don't be shy with the knob. Position yourself in front of the microphone behind the pop filter, about 4 to 6 inches or so from the capsule, and speak at a normal volume level while watching the green LED at the bottom of the input channel strip labeled"-20."Alternate between adjusting the gain knob and speaking into the mic until you see a good amount of bright, healthy flickering on that green LED.

And as you do this, also watch the red LED labeled"-20." OL, which stands for overload. If this red light lights up, it means your gain is too high. So a good approach is to bring the gain knob up little by little and keep checking the mic until you see your first flicker of red, and then dial back on the gain by a small but significant amount, and then you're in a pretty good spot. Once you've set this gain stage, route the mic channel to groups 3-4. And since we're dealing with a monophonic signal, it's sort of redundant to route it to both groups 3 and 4, so take the pan knob, turn it all the way right, so the signal is only going to group 4. Then bring the input fader and the group 4 fader both up to unity.

At this point, it's not really a bad idea to monitor the level of the signal that is leaving the mixer on group 4. To do this, Engage the button labeled subs 3-4 on the output section of the mixer in the little white rectangle labeled CR slash phones source. And then as you talk into the microphone, you will see the output level appear on the LED meter display on the right-hand side of the mixer. Of course, it's only going to be on the right side because you've panned this channel hard right.

What you want to see is good, healthy activity, peaking somewhere in the sort of upper green LED area. Very occasional yellow is okay. Don't feel like you absolutely need to be hitting that yellow LED.

It's not the worst. But you definitely do not want to be lighting up the red LEDs. So again, if you see them, back down on the input gain if necessary.

Now, in case you're curious, I'm going to show you how to do this. CR slash phones stands for control room slash headphones. And essentially, this part of the mixer allows an engineer to monitor some signal other than the main mix during a live session. And if you really want to know more, check out the manual, which is available on the course website. The relevant section starts on page 19. Anyway, when you're done monitoring the output level, disengage the subs 3-4 button.

Now... Over at the patch bay, using one of the longer patch cables, plug one end into group 4 mixer output and plug the other end into the first line input on the Focusrite interface, keeping in mind that this connection is a combo jack on the front panel of the Focusrite interface itself, not actually on the patch bay. On the Focusrite front panel, bring the input 1 gain knob to a mid-range position, around 5 or so, and again... Position yourself close to the mic, speak at a normal level while watching the LED input meters on the front panel of the Focusrite, and adjust the Focusrite input gain knob as necessary.

If you see the signal very occasionally light up yellow-orange, that's okay, but you should not see any red. If you do, dial back on the Focusrite input knob little by little until the red completely disappears. Step four, this is a quick one, configure the computer for recording. For this project, you're going to use Audacity. In Audacity's Preferences panel, go to the Devices section.

Make sure the correct audio device is selected for playback and recording, and make sure the number of recording channels is set to 1 mono. Also, click over to the Recording section and make sure Software Playthrough of Input is checked. Step 5, monitor with headphones.

You are in a closed space with a live mic, so it's probably, no, definitely a bad idea to monitor the signal with loudspeakers. The reason is that the microphone will capture the sound coming out of the loudspeakers and the system will feed back on itself. It'll get very, very loud very quickly and you'll be a very sad person.

So instead, I want you to monitor. With headphones, to do this, grab the headphones and plug them into the second headphones jack on the Focusrite interface. Now, please, please, please, whatever you do, do not lose the little quarter-inch adapter that clips onto the headphones plug. They are so easy to lose, and we don't have many, so do everything you can to make sure that adapter stays on the headphones.

With headphones plugged in, turn that headphones knob to a mid-range level to start. You can adjust as you go. In Audacity, turn on input monitoring by clicking the small mic icon near the level meters and choosing start monitoring. Once you do this, you should be able to hear yourself in the headphones as you speak into the mic. If you don't hear yourself, it's possible that the Focusrite output configuration has changed.

So in that case, open up the Focusrite control app, which should be in the dock. And on the left-hand column on the output routing page, make sure... Playback 1-2 is assigned to line outputs 9-10, which is what the focus right refers to for its second headphones output.

As you monitor the signal, there might be a small amount of latency, which is a short delay between saying something and hearing it. But in this case, a little bit of latency is normal and to be expected. Once again, speak into the mic and watch the level meters, this time on the Audacity window. And once again, monitor those meters.

You should see plenty of green activity with the signal peaks maybe reaching into yellow-orange territory, but make sure the meters do not reach all the way up to 0. If they do, the tiny notch at the end of the meters will turn red and stay red, and to turn those off, you'll need to click on them to acknowledge that there has been an overload. Okay, at this point, pause and take a photo of the patch bay, the mixer, and also a zoomed-out shot of your mic setup so that the entire microphone and microphone stand are visible. Name these three files with the following names and set them aside because you're going to include them in your submission.

Step six, record. You're now ready. to record the sentence you saw at the beginning of this video. But before you do, a couple of tips to keep in mind.

Number one, try to maintain a fixed distance from the mic as you speak. This will help keep the overall level of the mic signal fairly constant. Number two, and I cannot stress this enough, speak the sentence slowly and clearly with...

plenty of silence between adjacent words because this is going to make it noticeably easier to chop them up later on and tip number three if the latency in the headphones is distracting that's totally understandable you can take off the headphones but i don't recommend letting them hang around your neck because in that position the sound coming out of the headphones might actually bleed into the mic giving you a not very good recording so here's that sentence again You can pause the video here if you like. Click the red record button in Audacity, which will automatically create a new mono audio track. And when you're done recording, click the stop button. Step 7. Listen back.

With headphones on, press spacebar to play back the recording. Or if you like, hold shift and press spacebar to loop playback. Adjust the headphones level on the Focusrite if necessary to reach a comfortable headphones level.

Now, I want to pause and take just a moment to point out that even though this is a monophonic recording, just one channel of audio, you're hearing sound in both ears. That's because Audacity and DAW software in general is in some ways a simulation of an analog mixer. And by that I mean, here we have a single track. You can imagine this as being equivalent to a single input strip on a mixer. And although the track itself is mono, the output of this track can be, and in this case is, stereophonic.

And just like on a mixer, there's a pan control over here on the left, and by default it's centered, which is why you hear that mono signal equally in both ears. And you can move this pan slider if you want, and you'll hear how it behaves. It's probably very predictable.

Just make sure to recenter it. when you're done okay so as your recording plays back listen to the sound and watch the waveform very closely giving it the full attention of your ears and eyes don't just listen to the words that you spoke but listen to the sound of the entire recording including the words the stuff between the words and anything else that might have snuck in there and assess the quality In particular, what you should be looking for are things like clipping, basically any part of the waveform that goes past the upper or lower boundaries of the track. So, for example, something that looks like this is going to require a do-over.

And you also want to make sure that the signal level is also not excessively low. So something that looks like this is also going to be problematic. Now, in this case, you've got a fairly poor signal-to-noise ratio, which means if you try to...

amplify this recording to a healthier level, you'll also be bringing up the ambient background noise that was also recorded. What you want is a nice, healthy-looking waveform, making good use of the available vertical space on the track, though don't feel like your waveform has to be completely maxed out or anything like that. The words should be loud and clear, and the level of the background noise from the room and the analog equipment should be, comparatively, extremely low.

Since you're using a pop filter, popping the mic is a pretty minimal concern, but sometimes a pop can sneak by, so listen really carefully for pops as well. If you're happy with your recording, you can move on to the next step, or if you think your recording could be improved, delete the track using the X in the upper left, and record it again as many times as you like. Step 8, preparing for editing.

First, let's do some initial cleanup and save the original recording as an audio file. And from here to the end of the video, I'll introduce some useful Audacity keyboard shortcuts along the way. First of all, you can zoom in on a region by clicking on it and then holding Command and pressing 1 repeatedly. You can zoom out with Command 3. You can fit the entire track on screen horizontally with... command F and you can fit it vertically with shift command F.

Now your recording may have a bit of silence at the beginning and the end so trim that off by clicking and dragging to select these regions and then hit delete on the keyboard to remove a selection. And you can just eyeball this it doesn't have to be super precise but make sure you don't actually cut into the waveform itself. Then in the File drop-down menu, choose Export Audio. Make sure AIFF signed 24-bit is selected as the file format.

Name it lastname underscore original and save it in your personal folder, which is in the Music 407 folder, which is on the desktop. You're going to include this file in your project submission, so don't lose track of it. Now, since we're done recording, now what I want you to do is monitor on loudspeakers.

instead of headphones. So the first thing to do, and this is very important, is mute the mic channel on the mixer to avoid potential feedback issues. And here, we're going to monitor through the mixer like we did in project one.

So route the first two outputs of the focus right into mixer inputs one and two, route groups one and two out from the mixer to the loudspeakers, and then configure the mixer appropriately, which includes routing buttons. pan knobs and the relevant faders keeping in mind that these group faders should be relatively low for a comfortable monitoring level so don't just zoom them right up to unity now at this point take another photo of the mixer and another of the patch bay name these two files like so and set them aside to be included with your project submission step 9 digital audio editing Here is where you're going to do a fair amount of work on your own. So in summary, you're going to cut, splice, and rearrange this recording so that it sounds like this phrase.

Hello, welcome to music 407. So this means you'll need to rummage through this recording, really up close and personal, locate the appropriate phonemes and syllables, and then build this new sentence piece by piece. Now... A couple of tips to guide you as you go. The first thing you need to keep in mind is that there is no way that your rearranged recording is going to sound like normal human speech.

So don't freak out if it sounds kind of funky. To be expected. But even though it won't be perfect, I do want you to try to be as accurate as you possibly can with identifying, extracting, and assembling the appropriate syllables within reason.

Tip number two. I strongly recommend you don't try to do all of your editing work within a single Audacity track or even within a single Audacity window. What I do recommend is creating an entirely new Audacity project window, locating individual audio chunks in the original file, and then copy pasting them one by one into this new project.

This way your original file remains intact and for this project I think this is just a really clean way of working. And tip number three. While your work will, to some extent, be evaluated on syllable accuracy, the technical quality of your work will be the more important criteria.

And this applies to things like the overall level of your recording, whether there's any clipping or mic pops. But in the context of editing, here, primarily, we are talking about clicks, which manifest as discontinuities in your edited waveform. And to show you what I'm talking about, Let's go ahead and assemble the first word, hello. Now the best candidate probably is going to come from this word, helpful, which is right here.

So let's zoom in, and then to play just this segment, you click on the beginning, and then hold Shift and click on the end of the segment of audio you want to hear. And then hit Spacebar, helpful. Now in some cases, it'll be visually. pretty obvious where one phoneme ends and the next begins, but this is not true in every case.

So you're going to have to use your eyes and your ears to do this. You can hold shift and click around to adjust your selection, and when you think you've got it, command C to copy, and then over in the other document, command V to paste it. So now I'm going to edit with intentionally bad technique, resulting in a click to show you precisely what you should avoid doing. So let's go find an O to complete the word hello. So a candidate for this might be the word no.

So just kind of roughly select something like this, Command-C to copy, O, and position the cursor appropriately in the other project, and then Command-V to paste. And let's play the whole thing. Hello, hello, hello. There's a big ugly click right in the middle of the word.

And if we zoom in on it, we are looking at a sloppy digital splice, where a compression is juxtaposed right up against a rarefaction. And this translates to a loudspeaker cone trying to move from out to in, instantaneously, and this produces. A loud clicking sound, and this is bad, and we should avoid this.

One option to fix this is to apply a very short fade in and fade out on either side of the splice. I'm talking very short here, on the order of like 1 to 2 milliseconds. So that works pretty nice.

Alternatively, you might be able to identify a good splice candidate. by locating a pair of what are called zero crossings. In other words, moments where the waveform crosses the zero axis.

For example, here's a moment where the waveform is transitioning from rarefaction to compression. And here, on the other side of the splice, there's a similar spot. So we can roughly select this region from crossing to crossing.

It doesn't have to be perfect. Just close enough. And then, here's the trick.

You press Z to adjust the boundaries to align with the nearest zero crossing. And then, simply hit Delete, and voila, should look and sound pretty nice. At least from a technical perspective. So, as you build this phrase, think of yourself as an audio surgeon, right?

You're going in. With your microscopic audio tools making little cuts here and there, trying to stitch the patient back up while keeping the scar as small and invisible as possible. And if you think this is tedious, do yourself a favor and imagine trying to do this in the 1950s with magnetic tape and a reel-to-reel tape machine using a splicing block, splicing tape, and a razor blade. And don't forget the waveform recorded onto tape is complete.

completely invisible to the naked eye. So your eyes will be of very little help, and you'll need to rely entirely on your ears. And this is to say nothing of the physical tedium of cutting and stringing together a bunch of tiny little pieces of tape. But that was the reality of electroacoustic music in the 1950s, which is pretty wild. Anyway, I'll leave the rest of the editing and rearranging to you.

But one last thing you'll probably want to do before you move on is to save the Audacity project containing the rearranged So go to File, Save Project. Note that this is different from saving as an audio file. In this case, you're saving the project itself, and you'll probably want to save this in your personal folder to avoid cluttering up the desktop or anything like that.

And just periodically save your work as you go. It's always a good habit. And finally, step 10, export and clean up. When you're finished, your rearranged file should sound reasonably accurate.

intelligible and your recording and editing technique should be of demonstrably good quality so no significant background noise no clipping no pops no clicks and no excess silence at the beginning or end of either file export the rearranged waveform as an AIFF signed 24-bit file just like the other one name it last name underscore edit and you're ready to submit You should have the following collection of files. Note that you do not have to submit the Audacity project file, just the images and the audio files. Now as you did in project one, put all of these files in a new folder called lastname__project2, control click and choose Compress to create a zip file, and submit the zip to the course website and stash your project2 folder in your personal folder. Then clean everything up and reset the entire studio which means de-patching the bay, disconnecting from loudspeakers first, working backwards, resetting the mixer, turning off the computer and the monitor and putting the microphone and its accessories back how and where you found them so in particular make sure you disconnect from the xlr panel first then disconnect the cable from the mic being gentle of course make sure to coil the xlr cable as neatly and gently as you can there is actually a technique here so don't make a tangled mess and do your best to imitate the technique that i'm showing here Don't just wrap it wildly around your elbow like a rope, please. And also collapse the mic stand so that it is completely folded up.

And that's it. Good luck. Feel free to email me or the course TA if you have any questions.