Producing Quality Audio Recordings – Part 2: Tips & Tricks
In Part 1 of this series, I talked about creating the ideal conditions for producing quality recordings. Now that we’re familiar with recording equipment and environment, we’re ready to talk about pre- and post-production measures—as well as tips to keep in mind when actually recording—that will make narrations smooth and easy to edit.
It’s essential, perhaps fundamental, to have a clean transcript—proper grammar, correct spelling, and logical paragraph breaks. The text needs to be large enough for the reader to comfortably see. As the transcript is a guide for speaking (not a text to be seen), it’s helpful if unfamiliar words have guidelines for phonetic pronunciation. For example, if the transcript mentions the term “MySQL,” you could write “[my S Q L]” or “[my see quill]” alongside it, depending on how you want the reader to say it. Taking these steps can help get the pronunciation right the first time around, and save time in the recording process.
Now that you have a clean transcript, it’s ready for the reader (whether it’s you or someone else) to have a go at. Ideally, a full dry read should be done—no recording—just to get a sense of the paragraphs and text flow. This also gives a chance to actually sound out any words that are difficult to pronounce. It’s helpful if the dry run simulates the actual read-through as much as possible.
Having done that, you’re ready to begin—but first, you’ll need recording software. You can go through the process more effectively by using an application that both records and edits audio files. Audacity is a free and useful program; you can download it here.
Once you’ve installed it, open it up to get started:
Useful recording techniques
At this point, we need to determine who the reader is. If it’s you, first ensure Audacity is using your desired input and output devices. Click the drop-down list to the right of the microphone icon and check to see which microphone Audacity is using. If the default is your computer’s internal mic, you might end up using that instead of an external mic. It’s a good idea to use a headset for playback, since it helps you hear the recording more clearly than your laptop’s speakers. Do the same thing with the drop-down list to the right of the speaker icon. If you neglect to do this before playing back the audio, you may notice that the sound is coming from your laptop’s speakers instead of the headset’s. As you read, take pauses between paragraphs. These gaps—which should be at least a few seconds long—give you a break and let you take a breath. If you stumble over a word, pause for a few seconds and then start again from the beginning of that sentence. This way, you can omit the mistake easily. Remember, you can always shorten or otherwise edit these gaps during post-production.
If someone else is reading, a tap on the shoulder is a quiet and effective prompt to begin. It’s much easier than trying to tease apart overlapping voices when editing (which can happen if you use audio cues). Even if the reader is comfortable with the reading after recording it once, it’s useful to ask them to do it one or two more times. Having multiple recordings to work with is ideal, since it gives you options during the editing process. For example, if the reader enunciates something in one recording more clearly than they did in the other, you can cut and paste across recordings to create a composite recording that draws from the best of both attempts. Of course, there are a number of factors—the reader’s age, your recording environment, the length of the reading—that may make this difficult, but it is certainly worth doing if possible. Warning the reader about your request in advance can make them more willing to cooperate when you make it.
If you’re working with young children, it’s helpful to keep a few things in mind. Children are less likely to get restless if you intersperse their recordings with play time. Encouragement can be a powerful tool; pauses can be opportunities to give positive feedback. If a child has little experience with narrating, it can be difficult for them to do it in a natural or conversational way—it tends to sound like they’re reading from a script. You could prime their delivery by having them imagine they’re talking to you or to a friend, instead of broadcasting to an anonymous audience. If their tone continues to be an issue, use a substitute. If you’re recording a boy’s voice, you may be able to use a woman instead. In television and film animation, women are often used to produce boy’s voices.
And of course, there are tips that are equally applicable regardless of who is reading. For headsets, lift the boom arm in front of the eyes and position the mic head at least three inches away from the face. If a desk mic will be used, you can comfortably position it at least five inches away from the mouth. These adjustments are intended to minimize the impact of sibilance, smacking, and other distracting sounds that can occur during recording, which will sound more prominent if the mic is placed directly in front of the mouth. You can take preemptive measures to prevent exterior noise when recording. Put cellphones on “silent mode.” Close all browser windows and any applications that may interrupt the recording, such as Skype or Outlook. If a noisy notification still makes its way into the recording, use the same strategy as the one for addressing mistakes—pause for a few seconds and start again from the beginning of that sentence.
Another thing to consider is company culture. If the door is closed, does that mean “knock before entering” or “leave me alone?” Putting a note on the door will tell any would-be knockers what you’re doing in that room. If the environment is a cubicle or open floor plan, see if there’s a spare office or if it is possible to borrow someone’s.
I’ve noticed that squeaky chairs will find a way to make noise even if someone is sitting perfectly still—that can be avoided by standing instead. Paper scripts have noise-making potential, whether stapled or loose. I would recommend that the script be open on the computer and read from there. Depending on the length of the script, it may be necessary to scroll down while recording, and that’s fine—the average mouse wheel makes hardly any noise when used.
And with that, it’s showtime. Click the button with the red record icon at the top to begin recording. To stop, click the button again or press the spacebar.
Okay, so now you’ve got a recording (or two) at your fingertips. It appears in Audacity as a waveform, which can seem intimidating. The first thing I would recommend is to check out a blog post I wrote earlier this year on some of Audacity’s basic editing features. That post deals with selecting and deleting an unwanted segment, adding silence to a clip, and making audible segments silent. That’s enough to do some quick and dirty editing, but you may be working with a recording that requires you to use more sophisticated techniques.
I always like to start my editing process by making sure the clip’s volume is just right. If you find that your clip is generally too soft or loud, you may want to use the “Amplify” function to fix that. You can access that tool by clicking “Effect” in the menu bar and then selecting “Amplify.” This will bring up a new menu with a slider. The farther you slide the indicator to the left, the softer your clip will be; dragging it to the right will make the clip louder. As with most functions in Audacity, you have the luxury of being able to preview your changes before actually applying them to the clip.
Great, now the volume issue is settled. But what about the reader’s pace? You may be working with a child or teenager who didn’t read as slowly as you’d like them to. You can fix this, too, by changing the clip’s tempo. You can find the menu for this feature by clicking “Effect” and then “Change Tempo.” You may notice that there’s also a “Change Speed” option right above that one—this may sound counterintuitive, but that’s not the best option. Changing a clip’s speed has the potential to alter its pitch as well. Changing the tempo, however, deals less with “speed” proper and more with pace. The slider menu for this feature functions in the same way as the Amplify tool—left is slow, right is fast.
The last feature I’m going to cover is one I’ve found very useful—noise removal. I’m not talking about occasional clicks and pops that happen when nobody’s talking. Removing those is a cakewalk—select the noise and click the “Silence Audio” button. The noise removal tool is intended for eliminating the kind of pesky, persistent noise—such as a constant hiss—that overlaps with narration, making it harder to tease apart. Noise like that can usually be avoided in the first place by using a good mic or headset, but depending on your recording environment, your device may only minimize the noise rather than eliminate it.
Fortunately, fixing it isn’t too difficult. The first thing is to find a brief moment in the recording where no one is speaking—the only audible thing should be the offending noise. Click the waveform and drag your mouse to select the segment. Play back just this selection to make sure there’s nothing there except the noise. Once you’re confident you’ve got a good noise profile (as it’s called), you’ll want to go to the “Effect” menu and click “Noise Removal.” Click the “Get Noise Profile” button. This will close the current menu. Now that the program knows which noise profile to work with, you’ll need to go back to the Noise Removal menu one more time. This time, just click OK. The program should remove (or begin removing, depending on the length of your recording) the problematic noise. A final note on the noise removal tool—it is going to remove whatever noise is contained in the profile. It will not, as I initially believed it would, sift through the entire recording and remove all noise, since it doesn’t know what you consider “noise” and can’t distinguish between that and the human voice.
Here is a brief video demonstration of these post-production techniques, taken from a webinar I recently gave on producing quality audio recordings:
We’ve now covered all the essential steps to producing high-quality narration—setting up your recording environment, selecting your recording equipment, pre-production tips, and post-production tricks. What advice do you have for us and for our readers? Please share it with us in the comments section below; we’d love to hear from you.
Contact our Learning Developers
Need to discuss developing e-learning? Creating curriculum for classroom training? Auditing and remediating e-learning for accessibility? Our learning developers would be glad to help.