Home Theater Forum and Systems banner

1 - 12 of 12 Posts

·
Registered
Joined
·
6 Posts
Discussion Starter · #1 ·
I found this site after a google search for an equalizer for my microphone. It brought me to the Equalizer APO page.

So I Started with the tutorial which told me to install REW V5. And then it got overwhelmingly complex. I generally avoid asking for help until I have spent some hours being hopelessly confused. :crying:

I want to just look at an equalizer and curve the line a little, and have that filter get applied to my real-time voice streaming.

I'm using a headset.

Can someone tell me how to accomplish that?
 

·
Banned
Joined
·
4,838 Posts
You can do that with the Equalizer APO you first looked at. The REW stuff is for those who want to equalize room & speakers.

Set up the equalizer APO, then you can adjust the EQ as you wish by ear.

And welcome to Home Theater Shack!
 

·
Registered
Joined
·
6 Posts
Discussion Starter · #3 ·
Thank you.

The preamp does work. But I need the visual equalizer to figure out how to set it up. I don't know what PK, PEQ, HP mean, or what frequencies to change the gain on. Also, if I do something like 50Hz gain -3dB, will that leave 49.9Hz alone or does it affect nearby frequencies too?
 

·
Super Moderator
Joined
·
9,280 Posts

Hi Jimmy,

What you are looking for is an equalizer. REW is not an equalizer. It generates frequency response plots of a speaker’s output in a room (among other things).


I don't know what PK, PEQ, HP mean...
PK is peak ( I think – someone correct me if I’m wrong), PEQ is parametric equalizer, HP is high pass filter (i.e., rolls out all the low frequencies).


...or what frequencies to change the gain on
If you can’t tell by ear generally what frequencies you need to equalize, that puts you at a real disadvantage. Unfortunately no one here will be able to make any recommendations without being able to hear what the mic sounds like. I’m going to hazard a guess that it doesn’t sound too good, since good headset mics aren’t cheap. Regardless, here are a couple of things you can do that will improve vocal quality in most situations, no matter what the mic:

1) High-pass the signal at about 200 Hz. This will eliminate any “bomminess” in the voice that might come through to the listeners on the other end.

2) EQ the sibilants for a natural sound – i.e., not so soft that it sounds like you have a pillow in front of your face, but not so “hot” that the “s” sounds sizzle. This could be done with a parametric equalizer set for ~6-7 kHz and a 1/3-octave bandwidth. Then adjust the gain as needed


. Also, if I do something like 50Hz gain -3dB, will that leave 49.9Hz alone or does it affect nearby frequencies too?
Equalizer filters affect above and below the center frequency; how much depends on how wide the filter is. For instance, when you hear about a 1/3-octave filter, it means the filter affects roughly 1/3-octave both above and below the center frequency. Ditto for 1/6-octave, 2/3-octave, 1-octave etc. filters. Here’s a graph of a filter to give you an idea.




Regards,
Wayne
 

·
Registered
Joined
·
6 Posts
Discussion Starter · #5 ·
Maybe this will work.

This is from a program called Voice Shaper, that should do what I want, but doesn't because it won't share the microphone with my voice chat program. If anything else touches an audio device, it won't work. (yes, I have exclusive mode turned off on my devices)


(if the image doesn't show up, you can see it here) http://i.imgur.com/gIhwJ0k.jpg
The bottom green line is the realtime input. The bar at the top are the adjustments I made to reduce the hiss and volume of high-pitched stuff. I'm not sure what the 2 yellow dots did, but they did something when I widened them out. It might have just scaled the graph while the blue line remained static.

Is there a way to manually translate that into the text format REW outputs?
 

·
Premium Member
Joined
·
1,400 Posts
Maybe this will work.

This is from a program called Voice Shaper, that should do what I want, but doesn't because it won't share the microphone with my voice chat program. If anything else touches an audio device, it won't work. (yes, I have exclusive mode turned off on my devices)


(if the image doesn't show up, you can see it here) http://i.imgur.com/gIhwJ0k.jpg
The bottom green line is the realtime input. The bar at the top are the adjustments I made to reduce the hiss and volume of high-pitched stuff. I'm not sure what the 2 yellow dots did, but they did something when I widened them out. It might have just scaled the graph while the blue line remained static.

Is there a way to manually translate that into the text format REW outputs?
Have you looked at the control panel dedicated to your computers on board sound card ? ( btw, it's a separate bit of software from the sound control panel for the computer itself )

Many of these sound cards have some EQing capability . ( My Dell's Realtek soundcard has a 10 band graphic EQ / only available for manipulation through the RealTek control panel ) .

:sn:

PS; thanks for the mention of that Voice Shaper software .
 

·
Registered
Joined
·
6 Posts
Discussion Starter · #7 ·
The soundcard software doesn't affect the usb mic. And the software on my laptop is really limited. Basically just has volume & balance.

Does anyone know how to set those points from the graph into REW's export format?
 

·
Banned
Joined
·
4,838 Posts
Here is a way to have REW give you those filter values.

In REW, you can start with a fake "flat" response measurement, here is the .mdat file for one View attachment VoiShap Fake Flat Measurement Plus Filter Set.mdat , then use your desired curve as a "house curve," as in this text file View attachment VoiShapTargetHouseCurve.txt , have the EQ function apply its wizardry, then hand-tweak the result a little, and you get the following result, pretty close to what you are looking for:

voice shaping filter set - target and predicted.jpg


The filter value file is here View attachment Voice Shaping Filter Set.req .

And here is a jpeg of what those values are. Had to throw in one LPQ filter at the high end, don't know if the utility will accept that filter type, hope so. Good luck.

voice shaping filter set - filter values.jpg
 

Attachments

·
Registered
Joined
·
6 Posts
Discussion Starter · #9 ·
thank you. That stuff confused me so much.

I tried it and it does indeed change the sound. I am still tweaking it, as I sound quieter to people. The graph I showed you had the line drop off at the lower end for some reason, but those frequencies were untouched, so I turned those filters off. And the top end as well.. the graph dropped down without me doing anything, but they don't seem to have any effect.

I have it almost perfect now. I have tweaked the gains, but do I have to do anything with Q? Googling Q doesn't come up with relevant pages.

Code:
Filter 7: ON  PK     Fc  8,707 Hz Gain -10.4 dB  Q  2.84
 

·
Banned
Joined
·
4,838 Posts
The term Q stands for filter Quality historically, and is the inverse of bandwidth.

Q = 1 / BW

So a high Q value is a narrow-band filter, low Q is wide band filter. The Q setting will affect how the different filter bands interact. You can use the REW project attached above to see how they act. Open the project in REW, open the EQ pane, open the EQ Filters box, and you can change any filter type from Auto to Manual and see overall result as you tweak away.
 

·
Plain ole user
Joined
·
11,121 Posts
More precisely, Q is inductive reactance per resistance. As you increase inductance or frequency, Q rises, and is an indication of how sharp a filter response is.
 
1 - 12 of 12 Posts
Top