• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required as is 20 years of participation in forums (not all true). There are daily reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Boominess in the female voice range?

Marth

Member
Joined
Sep 22, 2021
Messages
36
Likes
10
Location
Deutschland
Hi Guys,

I am a frequent reader of the Forum but just created an account because I ran into problem treating my listening room.

Problem: There is a specific frequency within the female voice range which is extremely position dependent. In my listenting spot the 'n' letters in a song are so extremely boomy that my eardrum hurt at even moderate SPL.

When I leave the "sweet spot" it is fine. My frequency SPL curve has 10dB variations in the mid range frequency so I suppose this is a very particular case and problem related to one frequency in that song.

My questions are as follows:

1. How can I figure out the exact frequency of that part of the song so that I can better adress it. Room EQ Wizard doesn't seem to record frequency over time, just SPL. Is there a good software for this? I do have the standard UMIK-1 from MiniDSP.

2. I doubt that this is only a raise in dB as it really sounds boomy. However I don't have any high reverb above 80Hz. Can boominess be caused by combfilters/SBIE as well?

Some information about my system:
- I have the Linkwitz LXmini. It's basically a point source from 70Hz onwards. Radiation pattern is monopol from 70Hz to roughly 700hz where it transitions into a pure dipol. Around the transition region of 700Hz the backward part of the diplo cancels out the backward portion of the monopol thus creating a cardioid radiation pattern. Below 70Hz there are two dipol subwoofers which are pointing towards the listenting spot but are not toed in.

Due to the diplol character I have no problems with sidewall or frontwall (behind speaker) reflections. All early reflections <40ms are caused by the backwall (will be treated soon) floor and ceiling.
Room is 4.55m x 4.65m x 2.5m and I do have the respective modes wrecking havoc on my low end at 40Hz and 70Hz. However I doubt that this is related to the boominess in that female singing voice range...

I really need help in finding those problematic frequencies.

In my sweetspot the Low end response is fine so I can't really move around much. At the spots where that singing voice boominess is gone the lowend booms like mad.

Best regards
Martin
 
OP
M

Marth

Member
Joined
Sep 22, 2021
Messages
36
Likes
10
Location
Deutschland
Follow up information:

It's from this song around the 1min16sec mark. The letters/pitch which make my eardrum shake and boom are marked below.

Anne Sofie von Otter & Elvis Costello


Lyrics
It's not open to discussion anymore
He's out again tonight and I'm alone once more
He's all I have worth waiting for
But baby plays around
And so it seems I've always been the last to know
To hold on to that girl, I had to let her go
I wish to God I didn't love her so
'Cos baby plays around
I try to be strong hold on to my pride
He doesn't even know it's wrong, how much
I hurt inside
And heaven knows I've tried
But baby plays around. Just a plaything
It's hard to reconcile the facts I'm facing
It's not open to discussion anymore
She walks those shiny streets
I walk the worn out floor
She's all I have worth living for
Baby plays, baby plays around
 

Hipper

Addicted to Fun and Learning
Joined
Jun 16, 2019
Messages
728
Likes
587
Location
Herts., England
What you could try is to play the particular song in the region of the boominess and use REW to measure the song (instead of measuring the REW signal). Look at the SPL graph and see if anything stands out in the frequencies.
 
OP
M

Marth

Member
Joined
Sep 22, 2021
Messages
36
Likes
10
Location
Deutschland
I actually used the SPL recorder and at the sweet spot those two pitches are 5dB higher compared to a spot where it is fine. However the recorder only record SPL over time. Can I record frequency response without using the built-in sweep?
 
OP
M

Marth

Member
Joined
Sep 22, 2021
Messages
36
Likes
10
Location
Deutschland
What you could try is to play the particular song in the region of the boominess and use REW to measure the song (instead of measuring the REW signal). Look at the SPL graph and see if anything stands out in the frequencies.
Thank you for the reply. Will check if that can be done in REW later today.
 

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,581
Likes
3,806
A much easier way to do it would be to open the file in iZotope RX and analyse the spectrum.

I can do it this evening if you send me a snippet containing the problematic sounds.


In a second time, you would need to run a 20Hz-20kHz sweep with REW and see if there is a bump at the frequency where the "N" occurs.
 
OP
M

Marth

Member
Joined
Sep 22, 2021
Messages
36
Likes
10
Location
Deutschland
A much easier way to do it would be to open the file in iZotope RX and analyse the spectrum.

I can do it this evening if you send me a snippet containing the problematic sounds.


In a second time, you would need to run a 20Hz-20kHz sweep with REW and see if there is a bump at the frequency where the "N" occurs.
Thank you! Should I record the song playing with my mic or simply find it on youtube with time stamp?
 

PierreV

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
1,354
Likes
4,342
There seems to be a significant bump at 350Hz on the first n
Listening to the track both at the sweep spot and a bit off doesn't really cause anything to stand out for me.
Recorded with VBaudiocable in wasapi exclusive at 24 bits 44.1kHz


1632312411075.png


1632312652964.png
 
OP
M

Marth

Member
Joined
Sep 22, 2021
Messages
36
Likes
10
Location
Deutschland
There seems to be a significant bump at 350Hz on the first n
Listening to the track both at the sweep spot and a bit off doesn't really cause anything to stand out for me.
Recorded with VBaudiocable in wasapi exclusive at 24 bits 44.1kHz


View attachment 154992

View attachment 154993
Wow Thank you very much for the analysis. 350Hz seems reasonable. I will check and post my room measurments later today when I am back at home. It appears to me that the recording engineer did some "magic" here? That peak at 350Hz appears too synthetic and clean without any roughness... But I have no experience in that field.
 
OP
M

Marth

Member
Joined
Sep 22, 2021
Messages
36
Likes
10
Location
Deutschland
There seems to be a significant bump at 350Hz on the first n
Listening to the track both at the sweep spot and a bit off doesn't really cause anything to stand out for me.
Recorded with VBaudiocable in wasapi exclusive at 24 bits 44.1kHz


View attachment 154992

View attachment 154993
Do you know a freeware which can do what you did here? Just interested in that frequency analysis.
 
OP
M

Marth

Member
Joined
Sep 22, 2021
Messages
36
Likes
10
Location
Deutschland
That went over my head... :) But I will dig into it.
350Hz isn't a frequency that's easy to treat. I don't want to use absorption in first reflection points as I don't want to change/alter the spectral information of the reflected sound.
 

PierreV

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
1,354
Likes
4,342
Wow Thank you very much for the analysis. 350Hz seems reasonable. I will check and post my room measurments later today when I am back at home. It appears to me that the recording engineer did some "magic" here? That peak at 350Hz appears too synthetic and clean without any roughness... But I have no experience in that field.

You can't really say from this very short sample imho (the selected area is in white). On my office speakers (LS50+SB16Ultra) everything sounds "normal" - on my ****** desk speakers, the N sounds OK, maybe a tad abrupt, but another place in the song resonates with my desk :) What bothers you may be just such an interference/cancellation in the room interaction. (I guess the Linkwitz doesn't resonate but I have no experience with them)
 
OP
M

Marth

Member
Joined
Sep 22, 2021
Messages
36
Likes
10
Location
Deutschland
You can't really say from this very short sample imho (the selected area is in white). On my office speakers (LS50+SB16Ultra) everything sounds "normal" - on my ****** desk speakers, the N sounds OK, maybe a tad abrupt, but another place in the song resonates with my desk :) What bothers you may be just such an interference/cancellation in the room interaction. (I guess the Linkwitz doesn't resonate but I have no experience with them)
It's the room for sure. I have little treatment so far. The speaker doesn't stimulate the room much due to the directivity and null planes present in the dipol or cardioid radiation patterns. I agree that it is likely a interference of some sort at that frequency. Ty!
 

daftcombo

Major Contributor
Forum Donor
Joined
Feb 5, 2019
Messages
3,581
Likes
3,806
Thank you! Should I record the song playing with my mic or simply find it on youtube with time stamp?
You can find the time stamp in Youtube, but opening the file in Audacity (freeware), cut a 20s snippet, save it in WAV and send it to me would allow the best quality.
 
OP
M

Marth

Member
Joined
Sep 22, 2021
Messages
36
Likes
10
Location
Deutschland
There seems to be a significant bump at 350Hz on the first n
Listening to the track both at the sweep spot and a bit off doesn't really cause anything to stand out for me.
Recorded with VBaudiocable in wasapi exclusive at 24 bits 44.1kHz


View attachment 154992

View attachment 154993
Here you can see that I have large variation around the 350Hz mark. This could very well be the problem... There are many more of course :)
Room1.jpg
 

Hipper

Addicted to Fun and Learning
Joined
Jun 16, 2019
Messages
728
Likes
587
Location
Herts., England

Hipper

Addicted to Fun and Learning
Joined
Jun 16, 2019
Messages
728
Likes
587
Location
Herts., England
Here you can see that I have large variation around the 350Hz mark. This could very well be the problem... There are many more of course :)View attachment 155035
If it is 250Hz that might be what is going on here where you have a lump ranging from say 240Hz to 329Hz. Can you somehow lower that by 5dB and listen.

There are two approaches - alter the room by DSP/EQ or speaker/listening position - or alter this specific track. If everything else sounds good and it is only this track that is the problem I'd alter the track. Forget any 'as the musician intended' thoughts - it must work for you. Just keep the original somewhere for another day.

If the track has only certain regions where this boominess occurs you could use Nero Wave Editor, a free download, to home in on the regions and lower the volume. You can target quite small regions.

https://www.nero.com/eng/downloads/?currency=gbp&vlang=gb&country=gb&

Scroll down to find it.

There may be software that can alter frequencies on tracks but I have no experience of them.
 
Top Bottom