Any suggestions on making my voice crisper?

Hey there, everyone! I’m completely new to recording, and I was planning to do something nice for an upcoming relative’s birthday party. Thing is, I have absolutely no idea about what I’m doing. Here’s the raw and edited test audio I made:

Raw Audio

Edited Audio

For the edited audio, I didn’t do much: I did Noise Removal → Compressor → Normalize, but it still doesn’t sound quite as crisp as the raw audio (indeed, it sounds a little muffled, at least to me), and I haven’t the foggiest idea on what to do. If anyone can figure out what to do, I’d welcome any help.

I’ll also include the technical details on the settings I used; again, I don’t know anything about audio editing, so the settings I used were from Google searches and Reddit:

Compressor:
Threshold -30 db
Noise Floor: -40 db
Ratio: 2:1
Attack Time: 0.10 secs
Release Time: 1.0 secs

Normalize:
Changed the peak amplitude to -3.0 db

Thanks in advance!

1 Like

Your raw audio has a very high level of noise, (fan ?).
Removing it will inevitably cause serious collateral-damage to the speech.
[before-after]
The raw audio also suffers from clipping.

I would re-record away from fans, e.g. via a smartphone in a car/garage/bedroom.
https://www.youtube.com/@BoothJunkie/search?query=booth%20improvised

I can’t listen because I’m at work.

Does compression actually help? A threshold of -30dB seems “drastic” to me. Don’t use compression if it doesn’t help. :wink:

Normalization is a linear volume adjustment. It doesn’t affect the sound character/quality. It’s sometimes a good idea to Normalize or Amplify for “maximized” 0dB peaks as the last step. (ACX has an “odd” requirement of -3dB or lower, but that’s only for audio books.)

Again… this is without listening…
Boosting the highs with treble boost or with the Graphic EQ effect can bring-out the “T” & “S” sounds, which sometimes helps with intelligibility. (It will also boost background hiss.)

Some (some more?) clipping (distortion) can add “grit” and bring-up the loudness or “intensity”. Normalize or Amplify for 0dB peaks first to get a “starting point”. The run the Limiter effect. Set it to Hard Clip with Make-up Gain. Try limiting to -6dB, and if you want more run it again.

(Note that if you run it a 2nd time without Make-up Gain, nothing will happen because it’s already limited to -6dB.)

If you just want to make it “louder” overall without adding distortion you can try the Limiter set to Hard Limit (Again with Make-up-Gain). Limiting is a kind fast dynamic compression.