Make narration warmer, maybe boost bass?

(UPDATE 2018-11-04: A consolidated posting of all the effects I ended up applying can be found in this post Voice improvement - consolidated effects.)

Hi, the attached has been equalized, normalized, de-clicked, and sped up 16 percent. Audacity has helped it sound much better than the original me :wink: However, it is part of nine hours of a course voice over, and I’d like it to be as “warm” and welcoming as possible to keep people over a sustained period.

The excerpt attached sounds a bit mechanical to me. I occasionally get comments it does not sound “engaging”. I’d really appreciate any expert advice on how to improve it. I’m thinking adding bass, but am not an Audacity expert, so not sure the best approach and settings. Any thoughts on adding bass or other improvements to help make this something people would be glad to listen to for a long time?

(As thanks for this and previous assistance, here are 12 free coupons for the current course on Udemy, as far as I know the leading online course on practical implementation of the standard PMI best practices: - please let me know if posting this is not appropriate.)

I’d make it smoother by compressing the dynamic range.
After compression, sibilance will be even louder so de-essing will be required.
The combination of compression then de-essing will make it sound more bassy …

Thanks. If I wanted to just lower my voice, make it deeper, what I call “adding bass” although I’m not sure that is the right description, can I do that?

Thanks, I found the answer, by some user named Trebor :slight_smile: here:

I found reducing pitch by 5 percent was perfect. No unnatural distortion, but adds exactly the kind of deeper voice that makes it particularly pleasant to listen to.

All best - Bill.

As a consequence of doing that there will be reduction of high frequencies, correctable with EQ …

Bass & treble boost.png


If you can afford the luxury of (pseudo)stereo soundtrack,
IMO that’s more interesting than plain mono for those listening on headphones …

Interesting, thanks.

I’m trying to play around with the Ozone Imager, however I cannot find any useful help documentation. Could be poor searching skills. What I have found on the web appears to be for different software than the plugin.

As best I can tell, there are two settings, Width, and Amount. Is that basically what I play with?

Do you have suggestions on settings for those to obtain nice narration enhancement?

Remember, you’re asking for opinions.

I just played your original clip way up at the top and it sounds like you’re trying to force-read your voice lower than it wants to go. Instead of a pleasant vocal range you’re missing the bottom few notes. It turns into vocal effects instead of voice tones. It’s your version of “vocal fry.”

'What project management DOES." “Does” is not in your normal voice range and that can be uncomfortable to listen to.

You might try reading in what would be your normal pitch range and then apply corrections.

Did you use Effect > Change Speed? That one causes the least digital sound damage, but it raises your voice pitch. I think you should just read faster in your normal voice. Yes, DeEssing is needed. You have slightly punchy SS sounds.

You can sometimes get a more ballsy announcer voice by getting closer to a directional microphone. Make good use of Proximity Effect. That can encourage P Popping, so push the microphone slightly to one side instead of reading direct-on. That can also make reading the script easier. I once played two different people by doing that. A man and a woman!

We also note “make me sound like an announcer” is a popular request. There is no effective formula which is why this thread seems to be a laundry list of effects and corrections. We also note that some corrections in this list are necessary to correct damage from other corrections.


Yes Ozone Imager only has two controls.
You can also mix the original mono track with the pseudo-stereo (dual-mono) version …

ozone imager on narration.gif
Just apply ozone imager to the dual-mono track, (to make it pseudo stereo).

All the threes: width=33.3 amount=3.3 & set the dual-mono track 3dB quieter than the mono, (just suggestions).

One more. Did you apply Noise Reduction? It’s not in your list. Noise Reduction can make a voice sound mechanical and “cell phony.” People are paying so much attention to the noise level they miss what it did to their voice.


Thanks both.

Hi, I’d like to try this. I have no idea what the settings should be. Are the ones in the pic in your post the recommendations, or otherwise can you help me decide what they should be?

Just prior to that post I mentioned de-essing, you should de-ess before boosting high-frequencies with an equalizer,
(or Audacity’s bass&treble), otherwise sibilant sounds will be too loud.

A quick & simple de-esser plugin for Audacity is here … Please help with de-sser. - #14 by steve

De-Ess suggested settings.png

Interesting. I also seem to have a de-esser installed with Audacity, as per this pic:

Is one preferable over another?

Also, re correction of reduction of high frequencies with Marvel GEQ, any advice on settings, or just use the ones in the pic you posted?


The one you’ve already got can do a more precise job, but it’s much slow.
The one I linked to above is much quicker.

The equalization settings will depend on the source material,
so it’s a case of trial & error with the sliders individually …

Marvel GEQ demo.gif