Voice Over Recording

Why Some Voice-Over Auditions Get Thrown Out Immediately

September 28, 2012 by Ken 2 Comments

Young caucasian business man making thumbs down If you’ve been a regular reader of this blog, then hopefully you know that your computer is an excellent home recording studio, an you can record professional voice-overs (as well as music and any other audio recording projects) right from home. If you are one of the many voice-over actors on our list who daily answer voice casting auditions at Voices.com (or anywhere else), you probably also know that you have to be persistent. I’ve found that as with any acting job, you have to do lots and lots of auditions for every job you land. To quote Tom Hanks from the movie A League of Their Own, “It’s the ‘hard’ that makes it good.” But that doesn’t mean you can’t improve the odds. It is helpful to know that certain kinds of auditions get immediately thrown out by a client. Of course, it would also help if those weren’t yours.

Why I (as a voice-talent hiring client) toss auditions

1. Non-custom audition

As someone who has both auditioned AND hired voice talent from Voices.com, I can tell you one thing for certain – sending demo reels with recordings that don’t answer the specific need of the client is one way to get your audition ignored completely. When, as a client, I say in the job description something like “I need a celebrity impersonation of President Obama,” and one of the auditions is a demo reel of cartoon voices and car commercials, I click the Stop button immediately and move on. Just take the extra time and record something that the client actually asked for. It’s fine to send a demo that you already have if it matches the client’s description, or if they don’t really give a description. Otherwise, leave the demo reel for your profile page or promo kit.

2. Poor recording quality

Another thing that causes auditions to get set aside is a poor quality recording. See our tips here for getting the best recording possible regardless of what gear you use (that series of articles also tells you all about the gear!) – How to Build a Home Recording Studio: Part 2 – Four Tips For Preventing Noise . It is so inexpensive, gear-wise these days, that there is really no excuse not to spring for the $75 or so for a decent USB mic at the very least – if you’re serious about making the voice-over thing into regular paying gig.
If you have the gear, but are just not happy with the quality of your sound, or time it takes to record demos and jobs, you can learn to fix both problems in our new home recording video tutorial course. Check it out here: The Newbies Guide to Audio Recording Awesomeness 2.
So now you know how not to have your auditions immediately thrown onto the garbage heap. If you’re keeping track of how often you land a voice-over gig, this should improve your job-to-audition ratio greatly.
Good luck!
For another take on why voice-over auditions get chucked, check out this article by Derek Chappell: http://thevoiceofyourbusiness.com/blog/2012/09/27/why-do-95-of-demos-get-tossed/

How to Build a Home Recording Studio: Part 2 – Four Tips For Preventing Noise

May 11, 2012 by Ken 4 Comments

This is part 2 in a series called How to Build a Home Recording Studio, where you learn how to set up a computer-based studio capable of recording pro-quality audio.
In part 1, I left off in the section talking about ways of preventing noise in our recordings. The best way to do this is to record in an acoustically ideal space. One that not only is dead quiet, but does not affect the audio in a negative way either.

What are my options?

Since those spaces are really hard to come by, the next best option would be something like a purpose-made vocal isolation booth. You need the kind treated with acoustic materials to prevent and/or absorb echos, allowing you to record only the signal. Let’s take a look at some of these.

You can buy a full-blown vocal booth, such as the Clearsonic IsoPac, which will run you about $1,100. You can also build your own, or convert a closet for the purpose.
I do NOT recommend trying to convert a closet in your room into a vocal booth. Yes, it will be quiet and block out external sounds. But simply blocking out the lawn mowers, trucks, and barking dogs is not enough. The room sound in even a tiny room can sound bad, especially in the case of a square or rectangular closet. Either stuff it full of very absorptive materials so no sound can reflect of the walls, or make sure there are no parallel surfaces – preferably both.
For most of us, it is often impractical or too expensive to have either a very quiet recording space, or a good sounding isolation booth. Most of us do our recording in a spare bedroom, and bedrooms are notoriously effective at producing “bad” echos that when added to the signal, make the audio worse.

Why do bedrooms sound so bad?

Sound bounces off of hard surfaces all over your room, combining with each other to amplify and/or reduce certain parts of the sound. Then all these different mutant versions of your voice arrive at the microphone along with the direct signal. The results are echo-y at best, and likely will also sound unnatural and funky (sounding muffled, tinny, or like it came over a telephone – or even all of these).
If you’ve watched a lot of internet videos where someone is narrating, you have almost certainly heard the echo-y room sound thing. In fact it is way too common that a very slick and professional looking video has poor audio laid over the top. Usually it sounds like the person is speaking in a bathroom or something.
In the case of talking-head videos (a person talking to the camera – not the 80s rock group:)), this is almost always caused by the fact that the narrator is relying on the built-in camera mic, which is several feet away. The further your sound source is from the mic, the more room sound will be recorded.

Alternatives to sound booths

One alternative to a full isolation vocal booth is a small and portable boundary or isolation “shield” that can be placed behind a microphone, such as the sE Electronics Reflexion Filter, or the Aurlex MudGuard Isloation Shield.
These have several different types of acoustic treatment materials to prevent your voice from bouncing off the walls, and helping to block any reflections from getting back into the microphone. These are usually semicircular and protect the areas behind and to the sides of a mic. But something like that won’t work for a talking-head video.
So the first thing we have to do is deal with room noise before any sound reaches the mic.
If you can afford acoustic absorbing foam on your walls, it will be so much easier to reduce room noise. A good place to start is with Auralex products. But if you do not have any acoustic foam on the walls, the following tips will help immensely.

If you don’t have any acoustic foam on the walls…

Tip #1: Use a mic with a cardioid pickup pattern.

Fortunately, most mics default to that pattern. Cardioid mics record only what is in front of them. They reject any sound coming from behind it and most of the sound coming from the side. These can help reduce room noise a lot.

Tip #2: Get your mouth close to the mic – like just a couple of inches.

This will help the mic get mostly your voice (the signal) and less of the reflected sound. This is probably the most important thing you can do if your recording space is less than ideal. This is doubly true if you are shooting “talking head” video. See our article, 3 Must-Know Tips For Getting Quality Audio From Your PC Recording Studio for more on this.

Tip #3: Make sure you record as loud as possible without distorting/clipping.

A LOT of people make this mistake. They record their voice at a low level, and then they raise the level after the fact. The problem is that they are also raising the noise when the do this. So make sure to capture as high a level as possible of your voice in the first place. You can use the gain level on your interface (if you have one) or use the software mixer panel controls to do this.

Tip #4: Noise Reduction.

The next step in producing clean audio is to reduce the noise that will inevitably be in our recording when our rooms are less than perfect. The way to do that is to use tools in recording software called noise reduction.
What NR does is sample a section of the audio that is ONLY noise (when there is no voice talking), so it knows what to turn down. Then the program separates the noise from the signal and gets rid of it, ideally leaving the signal/voice unaffected. That last bit is really hard to do. There is often some “artifact” left behind after NR is performed. It usually sounds like swirling water. So you have to play with the settings to find the right balance of noise reduction without making the signal sound too weird. And the more noise there is to start with, the worse that swirly artifact is.
If you do all four of those things, you can get the best possible audio out of the cheapest possible gear. It’s what I did in the $5.00 vs $500 thing.
That is the first answer to the question of how we move toward pro audio from the $5.00 setup – having a basic understanding of how to limit noise.
In the next installment of this series, we’ll talk about digital audio what kinds of microphones can be used to give us even better sounding audio.

3 Tips For Recording Stereo Voice

April 10, 2012 by Ken Leave a Comment

There are several ways you can create a stereo effect when recording a voice, like a lead vocal in a song or as an effect in a voice over production. But first, let’s talk about why you might want to do this.
A clean and clear solo voice is the most natural and usually the best way to produce a vocal track. But there are times when you might want to add spaciousness or thicken up the sound of a voice. This effect is used a lot in music recording, often to create an intensity at certain points (like before the chorus) in a song, or even as the primary sound of a particular singer or band. Jon Anderson, from the group Yes, is an example of the latter.
I’ve been listening to The Beatles a lot lately and it seems a huge majority of their songs have the lead vocal coming at you in some sort of stereo, or at least doubled way. By “stereo,” I mean that the part (whether that be a voice or anything else) is actually two versions of itself and sounds as if it is coming at you from various directions as it does in real life. In the case of The Beatles, most of their recordings were produced in mono, so technically the vocal doubling wasn’t stereo. But the techniques they used to get thicker and spacey voices do produce a stereo effect if each of the signals is panned so that one comes from the left channel and the other from the right. In fact, that is exactly what happened when Beatles songs were remixed for stereo versions.
So what are these 3 tips for recording a stereo vocal?

Use 2 Mics When Recording

Of course the most obvious way to present a sound in stereo is to record it that way. Use two microphones to record one voice. This could be done using a stereo microphone (which is really two microphones in one), or two separate mics. Then in order to create the stereo effect, the output of each mic must go to its own track when recorded they can be panned left and right. If they stay together in the center of the mix (not panned at all) the result is a thicker (for lack of a better word) sounding voice like in the Beatles’ mono recordings. This is an effect all its own, called doubling, that is very common these days, but not technically stereo.
Anyway, using two mics for a stereo effect requires one more thing, which is to create a difference between the two recorded signals. This is because two identical signals, even if panned left and right, will still sound like only one un-panned signal coming from in front of you. In real life we take in audio through our ears (yeah, I know – alert the media), which as you have probably noticed, are on opposite sides of our head. Also in real life, sound comes to our ears only after it has bounced off of walls, floors, ceilings, and other surfaces. Each ear receives a slightly different version of the sound, allowing us to determine which direction the sound came from.
So in order to provide our listeners with a slightly different version of the voice for each ear, we have to make sure each mic points at the signal at a different angle. In a stereo mic, each of the two mic capsules point (usually) across the singer’s mouth instead of at it, almost as if the mics were aimed at imaginary people to the left and right of the singer. These two different angles provide the differences we need.
You can do the same thing with two separate mics if you make sure to put them really close together. But another way is simply to separate the mics by a few feet, and then unlike the previous method, point each mic directly at the singer’s mouth. Again, each mic will pick up the voice with different directional information and allow for a stereo sound when panned in the mix.

Sing Twice

The second way is to record the singer singing the part onto one track with one mic, and then record the singer again, singing it a second time (typically using the same mic) and recording this version onto a different track from the first. The differences in performance (subtle timing, breathing, phrasing, differences, etc.) will be more than sufficient for a stereo effect when panned in the mix. The Beatles used this method a lot in the early years prior to 1966.
One drawback of this method is that it sometimes singers don’t like it. John Lennon was particularly critical of doing things this way. He disliked it so much that EMI engineer Ken Townsend invented a way to create a similar effect without requiring the singer to sing twice. This technique came to be known as automatic double-tracking and common in modern recording. Essentially this technique electronically creates a second copy of the voice (or whatever is being recorded) and introduces changes to it that provide the required difference for the stereo effect. And that brings us to the third tip.

Copy The Voice

You can create stereo even if you only have one recorded performance. In today’s computer recording world it is simply a matter of copying and pasting like in a word processor. In Reaper, for example, you can click and drag an audio item while holding down the “ctrl” key and instantly create a copy in a second track. But that isn’t enough is it? Remember that you need to make them different from each other somehow, and then you need to pan them left and right to get stereo. Two copies of the same thing, regardless of how you pan them, is really just the same thing, only louder.
Perhaps the most common way to create the difference we need is to time-shift one of the versions by a slight amount. In most recording programs, simply dragging the audio item left or right in its track will shift it in time. Obviously if you move one of the tracks too much, it will be less stereo, and more of an echo. We’re only looking for a time difference 25 t0 35 milliseconds. It is a fascinating trait of human audio perception that we can hear two things separated in time by that amount and not consciously detect that we are hearing two things. But that small time difference is enough for our unconscious minds to hear a stereo effect if the two tracks are panned left and right.
Besides time-shifting other differences can be introduced, such as timber and tonal alterations to further increase the intensity of the effect. This is taking advantage of something called “The Haas Effect,” which you can read about in our article What Is The Haas Effect?
Use any or a combination of these three techniques for creating a cool stereo effect for a voice – or anything you’re recording for that matter.

ReaFIR Madness – The Hidden Noise Reduction Tool in Reaper

April 3, 2012 by Ken 13 Comments

Reaper audio recording software is awesome, and is what I recommend to everyone. I use it every day. And I also seem to learn something new about it every day. It has an incredible noise reduction tool built right in that I just discovered recently. The reason I didn’t know about it sooner is primarily due to the fact that reducing recorded noise is not its only or primary function. It just happens to do it REALLY well, and non-destructively (if that’s not a word already, I hereby create it). The tool is called ReaFIR. I know! Who’d have thought to look for noise reduction editing chops under a moniker like that?

Review of What Noise Reduction Does

First though, let’s do a quick review of what noise reduction does for us in the world of recording. There is almost always a bit of background noise in recorded audio. It is especially noticeable in recordings of just one thing, like a single voice. The noise usually comes from a combination of stuff happening in the space/room where the recording takes place, and the electronics of the microphone and other gear involved. Recordings sound much better if you can reduce the noise, and that is what noise reduction tools are designed to do.

But since the noise and the voice are both together in the same recording, it isn’t a perfect process. What noise reduction tools try to do is filter and removing as much noise as it can without also removing too much of the voice. In order to do this, the software has to know what noise looks like so it can separate it from the signal (voice). So you have to highlight a section of the recording where there is ONLY noise, and no voice, and feed that sample to the noise reduction tool. Once it has the noise profile, it can do its thing.

Enter ReaFIR

Reaper names its effect plugins by using “Rea” as a prefix (for Reaper), and then the name of the function after it. So ReaFIR means Reaper FIR. So what does FIR mean? It stands for Finite Impulse Response. If you would really like to geek out on what it really means, including all the scary math involved, you can read about it at the Finite Impulse Response Wikipedia page. But the simpler definition for ReaFIR itself from the Reaper manual is:
ReaFIR is a EQ and dynamics plug-in that includes an FFT spectrum analysis window. Amongst other things it can be used as a precision EQ, a gate, a fast attack/release precision compressor, a noise reduction tool.
Obviously it is that last thing we’re interested in. In the video above, I walk you through how to use ReaFIR to reduce the hiss noise that happened when I recorded a short voice over. BTW, this video is and excerpt from our new (to be released on May 1st, 2012) recording tutorial course, The Newbies Guide To Audio Recording Awesomeness – Part 2.

How Does It Work?

So see it in action in the video above. But here is is a written summary of how it works. Once you have recorded some audio onto a track, click the FX button in the track control panel. Then select VST: ReaFIR from the Cockos collection of FX plug-ins.
Next, go ahead and click the “OK” button and then close the ReaFIR window by clicking on the red X in the upper right corner.
Now just highlight/select a section of the voice-over audio that has no voice actually saying anything. In other words, pick a small area that was only noise. Then click on the Toggle Repeat button (down by the Play and Stop buttons). This is important because it will prevent any of the actual voice signal being played when sampling the noise for ReaFIR. With that area still highlighted, click on the FX button again to open the ReaFIR window.
In the Mode window of the ReaFIR tool, select Subtract, and put a tick into the Automatically build noise profile box. Now hit the Space bar on your keyboard (which will play the audio, which is just the noise at this point). You’ll hear it for a second and then it will be gone. That’s ReaFIR getting rid of the noise. Now hit the Space bar again to stop playback.
The next part is very important! Uncheck the box that says Automatically build noise profile. Otherwise ReaFIR will try to grab your voice and anything else on the track, and treat it as noise. Once you’ve done that, go ahead and close the ReaFIR window. Now all you have to do is click on the Toggle Repeat button again to turn it off, and listen to the entire track.
Voila! The noise is gone and only the voice remains. Pretty darned cool huh? And I think the result is better than a lot of tools dedicated to removing noise. I didn’t hear much, if any, of the weirdness that sometimes happens after noise reduction. The audio takes on a kind of swirly, under-water-y artifact sometimes. But I didn’t hear that in this example. However, there noise we got rid of was a low-level and consistent hiss type of noise. The louder the noise and the more varied the noise (if it contains lots of frequencies and intermittent clicks, pops, etc.), the more likely you are to have that swirly artifact left over after noise reduction.
So now you know about the secret noise reduction tool at your disposal if you use Reaper. Use it wisely.

How To Quickly and Accurately Cut, Copy or Paste Parts of a Song

January 23, 2012 by Ken 2 Comments

Cut Copy Paste Song Parts Whether you are recording your own music or working with a music file that already exists (usually a royalty free music clip for background music), there are many times when it is useful to cut, copy, and maybe paste sections of a song. Here are just a few of the more common reasons for needing to do that, just off the top of my head:

Shorten or lengthen a song to make it fit a presentation, video or voice-over production
Building a song by copying repetitive sections (entire bars or sets of bars) and pasting them
Fixing a problem in a song you are recording (for me it’s usually guitar buzzing I use this for) by clipping the same chunk of music from another part of the song and replacing the problem section.

Snap Tool in Reaper When working with music files, it is important that any edits you make do not mess with the timing. Otherwise it will be jarring to the listener. If you want to slice up a song, say, to make it fit a video, you want it to sound seamless in terms of the beat of the song. One excellent tool for doing this is the snap tool. This is a pretty common tool for editing both audio and video. In the picture on the left is where to find the tool in Reaper.

Snap To

The idea of snapping is simple enough. It is a way to ensure the edges of an item can only be be moved and placed so that they align with a grid. In the case of audio, that grid is frequently bars and beats, though you can change the grid to be other things, like minutes and seconds, etc. For our purposes in this article, we want to make sure we use bars and beats. In Reaper (see picture on the left), simply right-mouse click in the band across the top of the screen to change the grid settings.
The next important thing is ensure the song is aligned with that grid. When you’re recording your own music, this is not an issue as long as you are recording to the metronome which you set in the song project settings. And I HIGHLY recommend that you do that if at all possible. It makes things much easier in the long run.
If you have imported a song already recorded, you will have to do a few things to get its beats on the grid of the audio software. You’ll have to do a little trial and error, first setting the correct time signature for the song, then using the metronome and experimenting with the beats-per-minute setting until the song matches. Ironically, you’ll want to make sure the snap tool is turned off while you do this, so you can freely drag the song on the track to line up the visible beats to the grid.

Fixing Example

Once the song is matched to the grid, you’re free to copy, cut and/or paste and have the result be musically correct. Let’s take an example of when you need to correct a mess-up in the middle of your song (like my guitar buzz).
1. Just go to another part of the song with the same guitar chord playing without a buzz, and select the beat or beats containing that section. Make absolutely certain that the snap tool is turned ON when you do this.
2. Next, you click in the selected section and right-mouse-click to open the drop-down menu and select “copy selected area of item.”
3. Now open a blank track underneath the song track (you can see this in action in the picture at the top left of the article). This isn’t strictly necessary, but I find it helps to line things up BEFORE pasting into the song, which can save you lots of time and headaches. Paste the bit you copied into the blank track right underneath the buzzing guitar part. You may have to drag it left or right to m splitting audio file ake sure you line it up just right. Again, the snap tool needs to be ON at this point.
4. Next, click on the song at the beginning of the beat that has the guitar buzz and hit the “S” key on your keyboard to insert a split in the song file.
paste audio section 5. Now drag the left edge of the section with the guitar buzz in it to the right, creating a blank space where you will insert the part you are pasting. Things should now look like the picture to the left.
6. Now all you have to do is drag the pasted bit from the audio you pasted in the spare track up into the blank space. Having the snap tool on will ensure it doesn’t slip or slide to either side when doing this, which is a common problem when you don’t have snapping enabled.
7. The final step here is to make sure things sound smooth and seamless. Turn the snapping tool off (yes, OFF) and drag the edges of the new part you just pasted in a little left and right. Since cross-fading is turned on automatically in Reaper, this will allow the sounds to blend a little more smoothly. Otherwise the abrupt implanting of a different section might sound a bit jarring, creating clicks or pops at the borders. Cross-fading will help eliminate that problem.
And that’s it! It may sound a bit involved, but the entire process only takes about 30 seconds, which is usually a lot faster than having to play the part again.
Now you can use the same steps to remove parts from a song. Just select (with snapping turned ON) the part you want gone, hit the delete button on your keyboard and it’s gone. Now you can drag the part on the right back to join up with the song part on the left (closing the hole in the audio). Of course, if you had ripple-editing turned on, the part on the right would automatically be pulled to the left to join the two parts up.
If you want to make a song longer, just select a part of the song you want to repeat, usually a full measure, and paste it at the end of the song. Do this as many times as you need to in order to make the song as long as you need it to be.
Hopefully this simple tip of using the snap tool will help you cut, copy and paste bits of songs to your heart’s content.