Help with equalizing audio

Hi,

I make nature videos on YouTube: http://www.youtube.com/user/DynamixWarePro but I realized I didn’t equalize the recorded voice audio on my videos and that was one reason some didn’t sound so good. When I started making videos I sounded boring (not because I didn’t EQ the audio) but recently I was given a Rode NT1-A and a Alesis i02 express USB interface and since I was using a Samson C10U before to do voice over recording, it is an improvement. However I do find the Rode NT1-A a bit harsh and a bit too bright at the high end, but not much I can about that for now.

I have recorded some audio and would like help on EQing it. I have tried myself for the past two weeks trying to get it right and have videos I would like to finish and upload and feeling a bit frustrated that I can’t finish them without getting audio right.

All I can get is either a more muffled sound, sound a bit tinny or a bit to much high end. If I had a choice, I’d get a better mic that wasn’t a rode, but I am stuck for now. Can anyone help get me an EQ setting that will at least make the recording sound as good as it can get for what I have? I have attached a sample of the last recording I made.

I used a acoustic foam shield on the microphone to try and cut some harshness from it and use a pop shield. I also have large acoustic foam panels placed around the area I use my microphone to reduce echo. I can record the attached sample again if necessary, without the foam shield on the microphone if that is limiting the sound at all as I know trying to EQ a bad recording isn’t a good start.

I also use a Rode VideoMic Pro on my Canon Legria HF G30 and need help with EQ settings for recorded audio on it, but have not recorded anything with it yet to upload a sample from but I can when I do a recording with it.

I know everyone’s voice won’t EQ the same and I know I am probably going to be the only one that knows what will sound nearest like my voice, but can’t seem to get this right on my own and I have read and watched videos on equalizing recordings. Any help would be greatly appreciated. Can anyone help?

This is kind of refreshing because 1: You’re not a complete beginner, and 2: You’re already doing a bunch of the stuff we were going to recommend anyway.

We warn you that when you post like this, Do Not Help Us! This needs to be a clean shoot with no effects, filters or processing. Nothing like trying to take your effects out before we put ours in.

Do you remember what MP3 specification you used?

As we go.

Koz

The background is suspiciously quiet. Did you put a noise gate on the sound before you posted it?
Koz

Am I going the right direction? The first five seconds are plain you and everything after that is corrected.
Koz

You should try another test without the foam stocking over the microphone. I think part of my correction was to make the foam go away. You should be doing that.
Koz

That sounds like a technically good recording of someone with a “sing-song” voice which has wide variations in pitch.
The “sing-song” quality is not correctable with equalization , it may be possible to reduce it with pitch-correction software (autotune). Attached is an “after” where all pitch variation has been removed : it’s extreme and robotic, but illustrates the idea of limiting pitch-variation …

So now you get to tell us what we’re correcting. I got ride of the harshness by suppressing tones around 3000. I “took the foam pad off” by boosting tones from 7000 and higher. This is what my equalizer looks like:
Screen shot 2013-08-13 at 10.07.27 PM.png

Hi,

Sorry for the late reply. I managed to get the recording equalized how I wanted it by re-recording a few samples in a room which was sound proofed with acoustic foam and testing the mic at different distances away from my mouth because it has a proximity effect. New to using condenser microphones and just took a while to get things right. This recording I posted was just a raw recording without being edited. I didn’t change anything with the background, the mic itself has only a 5db self noise, so is quite quiet.

All I can get is either a more muffled sound, sound a bit tinny or a bit to much high end.

When I mentioned this, what I really meant was that the mic sounded a little bright as the Rode NT1-A is tweaked at 10khz for a bright sound and trying to eq a clear sound from it was making it sound a little harsh at the higher frequencies and was trying to avoid that as much as possible without making my voice sound a bit flatter.

… different distances away from my mouth because it has a proximity effect. New to using condenser microphones …

It’s not the condenser microphone. Directional microphones have proximity effect due to the vents, channels and tricks needed to make the directional pattern(s).

Were you able to get any set of filters which make you sound OK? I know people that use the NT1 and they all seem to be happy. “Harsh” tends to happen around 3KHz. That’s where your ear starts to bleed. Also known as “kids screaming on a jet.”

Koz
Screen shot 2013-09-09 at 7.11.52 PM.png

Hi,

It’s not the condenser microphone. Directional microphones have proximity effect due to the vents, channels and tricks needed to make the directional pattern(s).

Well the recording sounds different the further or close you are to the mic, not just the volume and Rode says the NT1-A has a proximity effect.

Were you able to get any set of filters which make you sound OK? I know people that use the NT1 and they all seem to be happy. "Harsh" tends to happen around 3KHz. That's where your ear starts to bleed. Also known as "kids screaming on a jet."

I started putting a lot of acoustic foam around the walls of where I record as I realized I had to to reduce echo from hard surfaces as trying to fix audio which has the echo ins’t the way to go. Once I finish the sound proofing, I try recording again. When I mentioned harsh, what I meant was that at higher frequencies the mic can distort a bit but I’ll re try my recordings after I do all this and see what happens.

The background is suspiciously quiet. Did you put a noise gate on the sound before you posted it?

When I made the original recording, I was using (I can’t remember the preamp, but it was USB) and kept the input gains low, the input volume low on Audacity and then all I did was use normalize and leveller after recording to get the right level of volume throughout the recording using the default settings for the normalize and leveller.

When I made the original recording, I was using (I can’t remember the preamp, but it was USB) and kept the input gains low, the input volume low on Audacity and then all I did was use normalize and leveller after recording to get the right level of volume throughout the recording using the default settings for the normalize and leveller.

I would not be using Leveler. Leveler is a brute force peak limiter and intentionally creates distortion. We are considering dropping leveler as the only use any of us have found for it is to create Air Traffic Controller voices.

“SoCal Approach. United three six Heavy clear for landing Two Four right.”
So unless you’re show is the controllers at LAX, that may be creating more problems than it’s solving.

You can try Effect > Compressor. That boosts low volume sounds at the expense of high volumes ones, making things louder (but bringing up the noise a little, too.) Use gently, that may solve your overall volume problems without creating harshness.

Amplify and its cousin Normalize just manage overall volume like turning up and down the volume on your radio. People find right away that they start running into peak volume damage using only those two. You can check for that with View > Show Clipping. Overload damage is shows by red stripes in the blue waves.

And if you just can’t make it all fit any other way, then yes, Leveler may be for you but I’d be happier with that if you weren’t complaining about harsh distortion.

And just because we’ve taken it this far, listen on somebody else’s sound system. I got a pair of terrible headphones once and I caught it before I tried mixing anything.

Koz

You can go a very long way by just suppressing the frequencies around 3KHz, the telephone frequencies. Try using the left hand dip on this pattern but return it to zero around 7KHz and flatten out the haystack bump on the right.

https://forum.audacityteam.org/t/help-with-equalizing-audio/30552/1

Stuff at 10KHz, if you can hear it at all tends to sound crisp and clear, not icepicks in your ears.

Koz

I still had the original curve.
Make the curve deeper – further down-- to increase the effect.

Koz
Screen shot 2013-09-19 at 8.48.35 PM.png

A de-esser plugin will selectively (dynamically) reduce areas with lots of high-frequency content [sibilance]

“SpitFish” de-esser

“LISP” de-esser

Sorry for a very late reply. I have put acoustic foam around the walls of the area I want to do my recordings in. I took off the foam cover on my NT1-A mic and did a recording. Here is a short bit of it. Its recorded at 24-bit 1 Channel Mono, 48Khz. It hasn’t been edited in any way.

I kept the input gain low so it wouldn’t pick up any other sounds other than my voice (it was recorded in a quite quite room anyway, I just noticed when I upped the gain, the self noise, maybe of the preamp, a Alesis i02 express USB, got worse, so tried to avoid that). There is a slight echo, I think because it picked up my voice coming of a small hard area of my desk I didn’t have foam covering properly.

I would not be using Leveler. Leveler is a brute force peak limiter and intentionally creates distortion. We are considering dropping leveler as the only use any of us have found for it is to create Air Traffic Controller voices.

I didn’t know that, I’ll avoid that in future. Relatively new to this, so have some learning to do. Thanks so far everyone for your help. I have the recording as best as I can get it on this project file, so can anyone help me equalizing and getting it to sound right?
NT1A_Test.aup (1.31 KB)

.aup files don’t contain any audio , you should post an audio file which either in WAV, FLAC format.
NB: I think the limit for an attachment here is 1Mb ,
so that translates to a 16 bit-depth WAV file of about 10 seconds @ a sample rate of 48KHz (mono, not stereo),
about twice that duration if you use FLAC .

For longer files consider uploading to a file-sharing website, or something like SoundCloud , and posting a link to the audio here.

Ok, here is a 24-bit FLAC 10 second clip of my audio recording.I think the reason I was getting distortion and my recordings didn’t sound good before is because I was using a leveler and doing other things before equalization and I wasn’t sure what to do with the audio to get the right volume levels before doing any EQ on it. This file is the unedited audio clip. I tested with Normalize then equalization and it sounded much better than I have had the audio before.
NT1A_Test.zip (714 KB)

It sounds good to me once normalized to 0dB to maximize the volume …
Normalize to ''0dB'' for maximum volume.gif
As for equalization my only suggestion would be to remove the infrasound, this does not produce an audible change but it makes the sound-wave more manageable if you were going to apply a noise-gate to attenuate the breath sounds between words, or apply dynamic-range-compression , or truncate silence to reduce the gap between words,


frequency analysis before-after the bass roll-off equalization shown.gif

I am new to Audacity and although I did get my book narration accepted and published with ACX specifications, it was not easy. I would love help finding the perfect equalization for my voice with ACX specifications. Can anyone assist?

Thank you in advance.

Shevon