What is the standard practice for recording audio?

Hi all,

I am hoping to get an idea of standard practice when it comes to recording audio to include with online learning. At the moment I use Audacity softwareand my laptop mic along with a makeshift sound booth to record and editthe audio which we link to projects we design in both Studio and Storyline.

The sound quality is not always the best as we have to record in any meeting rooms that are spare (no dedicated studio unfortunately) and I havebeen shopping around for portable sound booths and mic's as well as revisiting the software used.

I am hoping to get a feel for what software and equipment everyone out there uses and has success with.

Many thanks

14 Replies
Michael Gallagher

I use a Zoom H2N microphone. It's a standalone mic with built in recording functionality. You record the audio directly to an SD card that is plugged into the mic. Later, you can copy the audio to your computer and edit in Audacity. I use the optional accessory pack which gives you some nice options. The nice thing with this approach is that you don't have any computer noise in your recordings and it's compact.

Andy Learning Specialist

Recording starts up front. The best sound editor can't fix poorly recorded audio. So you definitely want a better recording mic than the laptops built in option. The sound booth is also a good addition. I also use audacity and have sound wave but don't use it much.

I have never seen an Elearning course that required such awesome production that it needed anything other than a decent mid-level mic and basic editing program. In fact, most courses have pretty bad audio and I know a lot of people that spend lots of money on audio devices and editing programs and still get bad quality due to poor execution. Unless you're making a film for the Sundance film festival, what you have plus a new recording mic should be fine.

Michael: The zoom is a good quality/value device although mine is very cheap feeling to the touch which turns a lot of people off. A mobile recorder like that isn't a bad idea if you don't want to spend hundreds of dollars on a mic.

Bruce Graham

Joe Waddington said:

I agree with Walter that good recordings start at the front end. Having the right tools and resources for the job make a difference.

I use a Snowball mic + Portabooth + Audacity for all recordings. A nice quiet area away from people traffic and noise is also a must.

...and knowing how to use Audacity "Noise Removal" tool.
Joe Waddington

Bruce Graham said:

Joe Waddington said:

I agree with Walter that good recordings start at the front end. Having the right tools and resources for the job make a difference.

I use a Snowball mic + Portabooth + Audacity for all recordings. A nice quiet area away from people traffic and noise is also a must.


...and knowing how to use Audacity "Noise Removal" tool.


Amen! White noise in narrations is like fingernails on a chalkboard for me!

OWEN HOLT

I have recently sacrificed quality for authenticity and flexibility with good results. I've found that working for a global company, people are used to hearing their colleagues through a phone or PC. Combine that with the fact that many of my SMEs are also scattered globally, I've resorted to setting up a VoIP internal call and setting Adobe Audition to record my sound card. The SMEs present as they would to a live audience and I record them. Because I'm capturing just the sound card (only the SME) I can coach them and give them direction without impacting the sound file being captured/created. I then use the recorded narration for my courses.

So far, participants appreciate the authentic feel of the audio. And while we are an "English speaking" company, this practice also lets me record alternate narrations in other languages fairly quickly and easily; catering to the needs of many of our users who appreciate the narration in their native language.

Long story short, I'm frequently using Audition and a USB headset/microphone combo.

Michael Harris

I came to e-learning narration by way of amusic recording studio background with a touch of general voice over tossed in for good measure. I set the same standards for recording a single voice (generally, mine) as for recording music: Pristine, CD-quality sound. I'm fortunate enough to have a relatively noise-free environment in which to record during normal business hours. I also have a well equipped and acoustically excellent home studio that is admittedly way overkill for e-learning narration. I won't go into the gory detals because it wouldn't be helpful. Suffice it to say there's a custom computer, inputs for 16 simultaneous sound sources, a fancy console with knobs and faders, professional monitors, and numerous microphones for various purposes.

The "standard practice" part of the question was the golden nugget. Coming at this from technical perspective, the standard practice is to faithfully capture an analog signal and convert it to a stream of bits. Realistically, any cardiod condenser microphone starting around the $100 retail price point will capturea signal well enough. Whether you like the way your voice sounds through a particular microphone is subjective. Try a few before deciding. The next piece, converting to a stream of bits, can be done poorly or brilliantly with little in between. Using the sound card built into a computer will invariably produce less than great sound. The alternative is a USB device, which is arguably the standard for high quality sound. I won't go into the growing choices of USB microphones. An alternative I like is using a good microphone with an XLR to USB converter. Blue makes one that works quite well.

So, we captured the sound and got a stream of bits. Now what? Here's a secret: All recording software does exactly the same thing in an indistinguishable manner. Faithfully recording a stream of bits is not hard, so the most expensive program will not do it any better than your favorite free program. The differences are mainly around workflow, multitracking capability, and bundled effects. Still, if we come at this from a technical perspective, some standards emerge. Be sure to record at 44.4 khz at a depth of 16 bits. That will establish CD-quality as your baseline. Plan on spending at least as much time editing audio as the actual recording.Get rid of extraneous sounds by cuts and sharp fades, not by using noise gates. Be judicious in the use of compression - audio recorded at a uniform level probably does not need any compression (or expansion, for that matter). Adjust the final level of your tracksto peak at 0 (zero) db. That will ensure the maximum possible volume and highest signal-to-noise ratio without distortion. Do your post-recording work with good headphones, not computer speakers. If you have not done this beforeyou will be surprised at what you hear. Some of your learners will experience your audio through headphones and you do not want them hearing weird stuff you didn't even realize existed.

After imporing your audio files into your favorite Articulate product, there are more choices to make. The file size versus sound quality trade-off never goes away. If it is in your control, publish several times at different bit rates and file compression settings and listen to each rendering through headphones. Choose the least aggressive compression settings that yield sound you are proud of. Do that by starting with no or minimal compression, successively compressing the audio until it is unacceptable and backing off to the last acceptable settings. Know that if the client specifies 16kbps MP3 files, the final product will be full of artifacts through no fault of yours.

Taking all of the above into consideration constitutes standard practice from a general audio engineering perspective. I think of e-learning narration as one of many applications for audio recording techniques. Regardless of the application, the basic practices for producing great sound do not change.

Mark Brown

I am not a sound engineer but I found a simple technique that let's me record my own voice fairly well without using a sound booth. None of the companies I've ever worked for would spring for a decent recording room. Most of my managers said 'Just use the mic on your laptop. That works fine.' Same thing with headphones: "Just get one from IT. They work fine." Even worse, the conference rooms have flat walls with terrible acoustics. "It's fine. We have to make due with what we have." Wrong, wrong, wrong.

First get a decent mic. I strongly suggest a Samson USB or something similar. It needs to be something you can hold in your hand. Do not get a Snowball, despite their excellent reputation. The problem with the Snowball or any mic (including the Samson if you put it in microphone stand) that sits on a table is that you'll get a "cold" sound that's not pleasant to hear. After much experimentation I found I could get around this effect by reducing the sensitivity of the mic through the standard Windows control panel down to about 30%, then holding the mic really close to my mouth. Now I get the nice warm sound I want my students to hear. Hold the mic about 1 inch next to your cheek, not directly in front of your mouth to avoid the dreaded S sound. If you watch professional singers perform you'll notice many of them hold their mics really, really close to their mouths.

Next get some decent headphones. Look for the word "monitor" in the description. I have some relatively inexpensive Sony's that work much better that the ones I got from IT. With these two pieces of equipment I can record and edit my own voice fairly well even while I'm sitting in my cube.

James Brown

Very well said Bruce!!.. Sometimes you have to play with this because you may remove too much.

Bruce Graham said:

Joe Waddington said:

I agree with Walter that good recordings start at the front end. Having the right tools and resources for the job make a difference.

I use a Snowball mic + Portabooth + Audacity for all recordings. A nice quiet area away from people traffic and noise is also a must.


...and knowing how to use Audacity "Noise Removal" tool.
Dwayne Beale

Mark Brown said:

I am not a sound engineer but I found a simple technique that let's me record my own voice fairly well without using a sound booth. None of the companies I've ever worked for would spring for a decent recording room. Most of my managers said 'Just use the mic on your laptop. That works fine.' Same thing with headphones: "Just get one from IT. They work fine." Even worse, the conference rooms have flat walls with terrible acoustics. "It's fine. We have to make due with what we have." Wrong, wrong, wrong.

First get a decent mic. I strongly suggest a Samson USB or something similar. It needs to be something you can hold in your hand. Do not get a Snowball, despite their excellent reputation. The problem with the Snowball or any mic (including the Samson if you put it in microphone stand) that sits on a table is that you'll get a "cold" sound that's not pleasant to hear. After much experimentation I found I could get around this effect by reducing the sensitivity of the mic through the standard Windows control panel down to about 30%, then holding the mic really close to my mouth. Now I get the nice warm sound I want my students to hear. Hold the mic about 1 inch next to your cheek, not directly in front of your mouth to avoid the dreaded S sound. If you watch professional singers perform you'll notice many of them hold their mics really, really close to their mouths.

Next get some decent headphones. Look for the word "monitor" in the description. I have some relatively inexpensive Sony's that work much better that the ones I got from IT. With these two pieces of equipment I can record and edit my own voice fairly well even while I'm sitting in my cube.



Thanks Mark, I was already looking at the Samson USB mic due to its portability so its good to know you have had success with it.

Sean Naes

I started with an AT2020usb plugged into my PC and used Audacity with satisfactory results. It all depends on your clients demands for sound quality, if they don't care then you have tons of leeway.

Here is the process that I recommend to get the best quality for the lowest amount of money. Under 150.00 if you already own a computer. (129.00 + tax for the Microphone, Audacity and Lame Encoder are free)


  1. Eliminate as much background noise as you can. Turn off all fans, air conditioner, heating etc

  2. Eliminate reflections (Bouncing sound waves that cause noise) by hanging blankets or other types of cloth that will absorb sound as opposed to reflecting it.

  3. Record about 20 seconds of silence then play it back and turn the volume all the way up to hear how much noise you have in your underlying track, is there anything else that you can hear? If there is and you can turn it off or eliminate it somehow, do so.

  4. Once you've cleaned up your recording environment as much as you can, your ready to record.

  5. Record your audio (let about 10 seconds of silence record at the beginning first)

  6. Once you have finished your recording your ready for processing.

  7. Select the 10 seconds of silence you recorded at the beginning using the selection tool, then click on EFFECT from the Top Menu then select NOISE REMOVAL from the pull down list. From the Noise removal dialogue, select GET NOISE PROFILE. (this will load the tracks underlying noise profile) NowPress + to select the whole track, thenthen click on EFFECTagain from the Top Menu then select NOISE REMOVAL again, this time the OK button should be active and click OK. This will now remove all of the frequencies in the noise profile that was loaded.

  8. Now you will need to go through and remove all of the mouth and breathing noises and anything else that you want to edit out. Once your are finished editing and are happy with your product its time to prep it for output.

  9. You whole audio track should still be selected, now you want to Normalize the volume (make it a consistent volume) Select EFFECT from the Top Menu then NORMALIZE from the pull down list. Select a value between -3dB and -6dB (I usually use -3dB) Then select OK. You will see the wave form visually change.

  10. Next I apply equalization to improve bass and treble. To do this, while your track is still selected click on EFFECT from the Top Menu then EQUALIZATION. From the Equalization dialogue I use the following settings Graphic EQ radio button selected, I use B-spline with a filter length of 4001. I then set the select curve pull down to BASS BOOST and click OK. I then repeat the process and Select TREBLE BOOST from the Select Curve pull down.

  11. Now to apply Compression to amplify the sound. Make sure your audio track is selected and Click on EFFECT from the Top Menu and select COMPRESSOR. At the Dynamic Range Compressor dialogue I use the following settings: Threshold -18dB, Noise Floor -50dB, Ratio 2:1, Attack Time .2 secs, Decay Time .1 secs and I have Make-up Gain for 0dB after compressing raido button selected then click OK.

  12. I then export the file in MP3 format using a bit rate of 128 kbps and a sample rate of 44.1khz. The export function is accessed from the FILE top Menu EXPORT selection in the pull down. If you have not installed the Lame Encoder, when you try to export in MP3 format you will be prompted to download it and install it. It's free.

Hopefully this will help you to produce the best quality audio that you can.

Never have I heard anyone say "I wish the quality of this audio was worse" but I have heard too many times "I wish the quality of this audio was better".