Best Practices for Narration Recording?


I am working on a project which is basically an audio book narration more or less. There will be music and effects added, but by others and not myself.

My job is to voice the narration. I’ve purchased a pretty nice mic/audio interface setup and have a room with (while not optimal) adequate soundproofing to record in.

But as an actor, recording settings are definitely NOT my expertise and I am struggling to get the levels and settings correct to produce a good sound.

my Mic is a Condenser mic which hooks up via XLR to the interface. I do have a shock mount and pop filter in use.

My particular mic has a toggle which allows you to either record at “normal volume” or to reduce the gain by -10db before the mic passes the sound on to the audio interface.

Obviously, the interface itself has a gain knob. Furthermore, as we all know, Audacity has its own “input volume” option. Plus more than likely some other volume options I am oblivious to or am just forgetting.

My question is this: What sort of levels do I want to have these set to? Is it better to record at a higher volume right from the input source? Or to record into Audacity with a relatively ‘quiet’ signal and then apply an Amplify/Normalize/Boost Effect of some sort?

My natural gut feeling is to turn the mic gain down -10 db via the toggle, but to then raise the audio interface gain relatively high. This way the mic is less likely to pick up ambient noise, and then the interface will boost that input to Audacity.

But when it comes to the Input Volume in Audacity, I just can’t seem to get a good level.

Clearly I don’t really know what I am doing and need some help getting my mind around this new frontier.

If anyone would be so kind as to educate me or point me in the right direction for the best practices for recording single voice narration, I would be very grateful.

Obviously, the interface itself has a gain knob.

It’s not at all obvious because you didn’t tell us what it was.

What is it? What’s the microphone? Who made them? Model numbers?

The -10dB switch is for playing a trumpet into your microphone. It helps avoid overload and distortion. You won’t need that under normal voice recording circumstances.

Slightly low recording is easily fixable in post production with Audacity or any editing package. Sound meter peaks in the -6 range. Too loud, overload, and clipping are hard or impossible to fix and usually fatal.

This is a perfectly recorded voice cut with no processing. It’s a test recording, so you wouldn’t have a side to side swing like that and the fourth segment is intentionally damaged, but overall, those are good target levels. Also note that when I stop talking, there’s little or no noise and very little room echo. That, too, is ideal.

The volume control in Audacity may or may not work depending on the answers to the model number questions.


The -10 dB toggle is usually to allow you to record very loud sounds without distortion. Unless you are screaming loudly down the microphone the switch should probably not be set to -10 dB.

We can probably give more specific suggestions when you’ve posted the details of your equipment.

Well, I was more so just asking for some ‘common practice’ type suggestions of how to get the best results, but you are right. Knowing the specs I am working with would be helpful.

I am using a Sterling Audio ST55 Microphone, which I plug into an Avid Recording Studio (AKA: M-Audio Fast Track II USB Audio Interface).

The audio interface is connected via USB to my laptop, which is running Windows 7 64bit (Intel Core 2 Duo, 4GB memory) and the latest version of Audacity Beta.

Hope this helps! Thanks very much.

-10 dB switch definitely set to off. You’ll get nowhere near 134 dB doing narration recording.

Ok, good to know!

In my particular case, much of the narration is indeed very ‘narrator-like’.

But this particular project is a dramatized reading. So there is a fair amount of actual character acting involved as well, with inflections and emotion.

Meaning that there may be times when voices may be raised, or the pitch may change pretty dramatically.

In some of my initial tests, despite having the overall gain settings quite low (at least so I thought), some of these more vibrant moments were clipping. And that’s been one of the bigger issues.

So how can I minimize the chance of that happening, whilst still keeping the volume input high enough to capture the ‘narrator’ style moments?

The microphone specifications claim that it can handle up to 134 dB without the -10 dB cut. That’s quite a lot louder than a revving chainsaw.

Setting the Pre-amp recording level:

  • Turn the Microphone Gain knob fully counter-clockwise.
  • Connect a microphone to the Microphone Input on the front panel using an XLR cable.
  • Set the Phantom Power switch to the “On” position.
  • While talking at your loudest levels, slowly turn the Mic Gain knob clockwise until the red clip indicator begins to illuminate. Then, turn the knob counter-clockwise until the clip indicator no longer illuminates.

Your microphone pre-amp is now set correctly.

Sorry I can’t help with Win 7, I don’t use it.

When you’re used to commercial recordings carefully processed, it can be an enormous shock the first time you make a live recording and find out how difficult it is to capture performances without damage.

There is another posting of someone performing the drum parts of a song by very close miking his mouth. The microphone he chose is correct, robust and very well behaved and will follow him wherever he wants to go, but the poor downstream electronics are all lying beat-up and bleeding by the side of the road.

It’s harder than it looks.

I don’t necessarily agree that theatrical recordings involve large changes in volume. They may do so on the stage, but not in radio or other commercial recordings. There are filters and processors you can use to even out the volume variations gracefully after you go for the full operatic expression thing while recording the show.

I need to go.