Existing users, log in.  New users, create a free account.  Lost password?


MacFixIt Logo
 
Contact Us | About MacFixIt | Who's Online  

Topic Options
#389332 - 03/29/07 10:51 PM "Fixing" Old People's Voices
dave393 Offline
New User

Registered: 08/21/06
Posts: 4
Loc: McLean, VA
I do a considerable amount of recording of narrations by very elderly people. The problem almost always is to enhance the quality of their voice, add depth and presence. I'm not sure I've done too well at that and would appreciate help here.

In addition, I recently worked with a quite elderly woman whose voice often rises into the stratosphere of frequencies. In short, she often is unbelievably shrill. I need to do something to "take the edge off" those piercing segments. I have both Amadeus Pro and Peak LE (with their plugins). I also have very little knowledge of EQ, compression, and all the rest. (When I've tried out EQ, I abandoned the effort. There are just too many sound parameters to keep track of when I'm searching for the right combination of settings.) Tips, detailed instructions, whatever, are strongly needed.

Dave

Top
#389333 - 03/30/07 05:20 AM Re: "Fixing" Old People's Voices [Re: dave393]
Al Bloom Offline
MacMaster

Registered: 01/20/01
Posts: 3472
Loc: Blacksburg, Virginia
Dave, have you posted this to the pros who hang around the Amadeus
forum?

web page

To your question, I believe the "Normalize" effect is what you seek for
the screecher.
_________________________
Al Bloom

MDD G4 867DP, OSX 10.4.11, OS9.2.2

Top
#389334 - 03/30/07 07:33 AM Re: "Fixing" Old People's Voices [Re: Al Bloom]
dave393 Offline
New User

Registered: 08/21/06
Posts: 4
Loc: McLean, VA
Al, thanks. Will post to that forum.

As for "normalize", as I understand it, it works by adjusting "gain" and not frequencies. Of course, cutting back on gain will dampen the shrill effect somewhat. But, at the cost of having to reduce the gain throughout. I don't think I'd want to do that.

Dave

Top
#389335 - 03/31/07 06:28 AM Re: "Fixing" Old People's Voices [Re: dave393]
Al Bloom Offline
MacMaster

Registered: 01/20/01
Posts: 3472
Loc: Blacksburg, Virginia
Section 5.3.2 (page 17) of my Amadeus Pro manual: Normalize evens things
out in one of two possible ways. Sort of like Sound Check in iTunes. Try it.
_________________________
Al Bloom

MDD G4 867DP, OSX 10.4.11, OS9.2.2

Top
#389336 - 03/31/07 09:31 AM Re: "Fixing" Old People's Voices [Re: Al Bloom]
dave393 Offline
New User

Registered: 08/21/06
Posts: 4
Loc: McLean, VA
Will do. Thanks again.

Dave

Top
#389337 - 04/24/07 06:03 AM Re: "Fixing" Old People's Voices [Re: dave393]
Gareth Williams Offline
New User

Registered: 04/24/07
Posts: 3
Hi Dave,

I personally wouldn't normalise the shrill lady's voice. All this does is to increase the gain so that the peaks are at the level you normalise to.

You basically have two options available to you as I can see. You must forgive me if appear to be condescending - I just want to make sure I cover the detail in full so I'll assume you know pretty little!

Firstly, to enhance the warmth/depth of a vocal it is helpful to compress the signal. Imagine a graph of a signal - time on the x-axis, level on the y-axis. All you are doing to the sound here is squashing the peaks of that line and lifting the troughs, hence reducing the dynamic range of the signal. This will make it appear smooother to your ears as there aren't such large fluctuations in level.

Secondly, for the shrill lady, you really can't do much other than EQ the frequencies which are the problem. There are three settings you need to know about when using a complex EQ plugin and you can probably safely ignore the rest and leave them set to default. 1 - Gain - the amount of cut or boost you are going to apply. 2 - Frequency - the frequency at which you will apply that cut or boost. 3 - Q - this is how wide or narrow that boost or cut is i.e. how much effect it has on the frequencies around the frequency you select to cut/boost.

For a shrill female vocal, firstly boost the gain and then 'sweep' through the frequencies to find where the shrillness really jumps out at you. Then change that boost to cut and the shrillness will disappear with it. If you find that by doing that you're losing a lot of top end, try adjusting the Q so that the cut works more locally to the specific frequency. Alternatively you could just use less cut so that you acheive a balance of not taking too much away but still reducing the shrillness.

I would imagine that you'll be looking for a frequency somewhere in the range of 1.5KHz to around 4KHz. However if it's false teeth whistle that's the problem I'd go a little higher up to about 6KHz.

Hope this is some help.

Top
#389338 - 04/24/07 02:52 PM Re: "Fixing" Old People's Voices [Re: Gareth Williams]
dave393 Offline
New User

Registered: 08/21/06
Posts: 4
Loc: McLean, VA
Thanks a lot Gareth. It was of considerable help. A Compressor and EQ seem the instruments of choice. However, I wonder how one makes the choice? I mean, in my scouting around the web I've picked up several Compressors and the same for EQs. So there they are. Sitting in the computer.

Without just trial-and-error usage, does a way exist to select the better from the pack of either Compressors or EQs? As I remember, the freebies I've picked up don't usually come with any sort of usage documentation and specs. Instead, if I scout the ones you pay for, are there characteristics to look for? Thanks.

Dave

Top



VersionTracker: Software Updates and Downloads | iPhone Atlas: iPhone Help, News, Tutorials, and Tips