How can I make my VoiceOver sound better


I have been developing VoiceOver videos and I am learning how to use Audacity to make good quality recordings. Can someone advise how I can improve the crispness and overall sound quality of the attached VoiceOver? Could you give me tips on what you did so I could learn and use it in the future? I have done the usual stuff you learn on YouTube, like filter curve EQ, compression, normalization. I have attached a sample of the recording.

That’s overprocessed. You have processing distortion sliding through the words.

Do you have the original WAV file?

If you didn’t announce to a WAV file or an Audacity Project, that’s the first problem. Never do production in MP3. MP3 gets its small, convenient sound files by shifting sound tones around and leaving some of them out.

Announce to this ten second test. Export as a WAV and don’t correct anything.


Sounds like compression has exaggerated the sibilance.
The cure is a de-esser plugin like desibilator

suggested  settings.png

Sounds like compression has exaggerated the sibilance.

You can get the same effect with excessive Noise Reduction. It might be good for us to sit in a corner and sort why effects do that. We know there’s a limit to what can be done with effects or the correction will take until next Wednesday. Nobody is going to wait that long.

The one I noticed was the honky trailing sounds such as “by” in “Embarrassed by.” at 12 seconds.

There’s plosive P Popping on “Depict” at 33 seconds. That’s microphone placement or acoustic pre-conditioning (blast filter).

We can sort all this out when you post your “clean” sample. We can’t take effects out of a sample to figure out what went wrong.

You know you’re in trouble when you start applying effects to correct the damage caused by your corrections. The other problem with piling on corrections is a never-ending cycle. We can fix the popping but that causes low pitch pumping. We can help that…


Thank you very much for helping me learn this and for taking time out of your day. Here is a short sample exported from AUP to WAV.

Excellent. Do you hear anything wrong with that? I don’t. It sounds natural and probably exactly like you. It has good tonal balance and I didn’t have any trouble mastering it to audiobook technical standards. I can listen to a class in that voice.

Can you post one where you take it all the way out to ten seconds? I want to hear “Produced by.” You had “P” sound problems in your first posting. If you do have those problems, there’s easy ways around them.

And pick up the pace or tempo a little. Sound like you’re trying to sell me milk. If you sell your services, you may get a client in sales. If you’re announcing your own classes, you don’t want to put anybody to sleep—unless that’s the goal.


This is Effect > Change Tempo. The First clip is 24% boosted. The second clip is natural you.

Don’t fall in love with that. Effect > Change Tempo can cause sound distortion. the goal is to announce faster in front of the microphone.

I stepped out of the engineering booth here. If you announce at the speed you want, then that’s the way it is. We can produce a good track either way.

I can post the audiobook mastering tools which can even out volume and expression and produce a standard loudness product. All without changing the quality of your voice.


What’s the microphone or recorder?


There is a little reverb from the room.
It’s at a low level where de-reverb plugins are worth a try …

The free version of Couture is sufficient, but ACON’s DeVerberate, which is better, is ~$80 :astonished:

There is a little reverb from the room.
de-reverb plugins are worth a try …

Or you can tighten up your voice with furniture moving pads.

Note there’s a pad folded over on the table.


I can second the Harbor Freight moving blankets. I have used them in the past when in a pinch and they do a good job dampening the room sound.

There’s even fuzzy rules for the blankets. Start with deadening the table (unless you’re using the Pressure Zone Technique), and then hit one opposing surface at a time. For example, table, then west wall, then south wall, and then start doubling up. Hit the north wall of the south/north pair, East wall in the east/west pair. Then the ceiling.

Note in the kitchen table studio, the wall behind the performer is the only surface not blanketed, and it’s possible to cover that one, too. As is posted, people make perfectly lovely pre-baked portable studios for purchase. You don’t have to do it for $25 at the hardware store.