[Bill Bremmer]

Hello,

I want to notch out a (noise) frequency of about 58.1 through 58.9 on Cool
Edit recordings.
How can I do this? Can I notch while recording or do I have to apply the
"fix" after I've recorded.

If this doesn't work, I want to filter any freqencies below 80 hz. Again, I
assume that it is better to filter while recording than after. How can I do
this?

Would filtering below 80 hz have much affect on a male voice?

Thanks,
Bill Bremmer

[Nil]

On 29 Jul 2007, "Bill Bremmer" wrote in
rec.audio.pro:

Can I notch while recording or do I have to apply the "fix" after
I've recorded.

None of Cool Edit's effects can be used while recording. They can
only be applied afterward.

[Don Pearce]

On Sun, 29 Jul 2007 11:21:41 -0400, "Bill Bremmer"
wrote:

Hello,

I want to notch out a (noise) frequency of about 58.1 through 58.9 on Cool
Edit recordings.
How can I do this? Can I notch while recording or do I have to apply the
"fix" after I've recorded.

THere is a good range of filters you can choose for this.

If this doesn't work, I want to filter any freqencies below 80 hz. Again, I
assume that it is better to filter while recording than after. How can I do
this?

Would filtering below 80 hz have much affect on a male voice?

That is a much better idea - in fact go for 90Hz rather than 80, it'll
be just fine. And this way you lose all the subsonic junk that just
makes for clutter in the mix.

You can't do this while recording - it is an "after" thing. No worse
for that, though.

d

--
Pearce Consulting
http://www.pearce.uk.com

[Paul Stamler]

"Bill Bremmer" wrote in message
. ..

I want to notch out a (noise) frequency of about 58.1 through 58.9 on Cool
Edit recordings.
How can I do this? Can I notch while recording or do I have to apply the
"fix" after I've recorded.

If this doesn't work, I want to filter any freqencies below 80 hz. Again,
I
assume that it is better to filter while recording than after. How can I
do
this?

Would filtering below 80 hz have much affect on a male voice?

Not unless it was a deep baritone or a bass.

Bill, it's probably a good idea to keep this discussion in the original
thread. But in any case...

In CoolEdit, unless your interface/mixer/preamp has a high-pass filter
option (which, as I recall, yours doesn't), you'll have to do the filtering
after recording. What version of CoolEdit are you using?

Peace,
Paul

[Bill Bremmer]

I'll try to keep one thread. But here is your answer. Cool Edit 2000 with
Studio Plug-In.
I've tried many of the filters supplied with Cool Edit (including making
notch filters of the noise frequencies) andhave had poor success. In the
instances the filter helped, the results were lost when I normalized. Note:
there is only about a 3 dB difference showing on my meter between the
original voice recording and just recording with the mic switched off.

I think the pre-amp cranked all the way up is producing the pre-amps own
system noise. Keep in mind that there is only about a 3dB difference on my
meter between the recorded voice and recording the noise with the mic switch
off.

What about using a condenser mike to get a stronger signal to noise ratio.
As I saidthe recorded signal only peaks at about 30%. Obviously I have to
normalize a lot to get a usable finished product.

"Paul Stamler" wrote in message
...
"Bill Bremmer" wrote in message
. ..

I want to notch out a (noise) frequency of about 58.1 through 58.9 on
Cool
Edit recordings.
How can I do this? Can I notch while recording or do I have to apply the
"fix" after I've recorded.

If this doesn't work, I want to filter any freqencies below 80 hz. Again,
I
assume that it is better to filter while recording than after. How can I
do
this?

Would filtering below 80 hz have much affect on a male voice?

Not unless it was a deep baritone or a bass.

Bill, it's probably a good idea to keep this discussion in the original
thread. But in any case...

In CoolEdit, unless your interface/mixer/preamp has a high-pass filter
option (which, as I recall, yours doesn't), you'll have to do the
filtering
after recording. What version of CoolEdit are you using?

Peace,
Paul

[Norbert Hahn]

"Bill Bremmer" wrote:

I think the pre-amp cranked all the way up is producing the pre-amps own
system noise. Keep in mind that there is only about a 3dB difference on my
meter between the recorded voice and recording the noise with the mic switch
off.

What kind of mic switch? A mute switch on the mic?
Anyway, your setup is broken. Either the shield of the mic cable is open,
or you have some ground loop in the system.
What external devices are connected to your PC?

What about using a condenser mike to get a stronger signal to noise ratio.
As I saidthe recorded signal only peaks at about 30%. Obviously I have to
normalize a lot to get a usable finished product.

A condensor mic may produce a 10 dB hotter signal - still not enough to
overcome a broken cable.

As you wrote that your recording suffer from some kind of static noise,
a simple low cut won't help you much. It will leave the higher frequencies
of the noise.

Do you get a clean recording when you set the gain to zero? Of course,
do don't record anything but white noise (hopefully).

What do you get recorded when you disconnect the mic from the mic amp,
with the gain setting to 4 o'clock?

Does the sound of the noise change when you touch the mic or the plug
or the cable?

Does the sound of the noise change when you move the mic around in the
room?

Norbert

[Paul Stamler]

"Bill Bremmer" wrote in message
. ..
I'll try to keep one thread. But here is your answer. Cool Edit 2000 with
Studio Plug-In.
I've tried many of the filters supplied with Cool Edit (including making
notch filters of the noise frequencies) andhave had poor success. In the
instances the filter helped, the results were lost when I normalized.
Note:
there is only about a 3 dB difference showing on my meter between the
original voice recording and just recording with the mic switched off.

I think the pre-amp cranked all the way up is producing the pre-amps own
system noise. Keep in mind that there is only about a 3dB difference on my
meter between the recorded voice and recording the noise with the mic
switch
off.

Well, now, you never mentioned *that*. Are you saying that the noise,
without normalizing, is only 3dB below the voice level? In which case it
must sound like you're talking at Niagara Falls. Or are you normalizing the
voice signal and separately normalizing the noise? Because that would
invalidate every measurement you've mentioned.

What does the noise spectrum look like when the mic is switched off? Same as
when it's on? Do the two measurements *without* normalizing; leave the gain
set where you had it (4:00), do a short test with the mic on, another with
the mic off, and don't touch the normalizing function. Don't talk. What are
the levels and frequencies?

What about using a condenser mike to get a stronger signal to noise ratio.
As I saidthe recorded signal only peaks at about 30%. Obviously I have to
normalize a lot to get a usable finished product.

You don't, really, but that's a very different issue. Let's leave it for the
moment.

Peace,
Paul

[Arny Krueger]

"Bill Bremmer" wrote in message

Hello,

I want to notch out a (noise) frequency of about 58.1
through 58.9 on Cool Edit recordings.

Use the DTMF filter.


Can I notch while recording or do I
have to apply the "fix" after I've recorded.

Fix the recording once it is made and saved. That way if you mess things up,
you just pull up the saved file.

If this doesn't work, I want to filter any freqencies
below 80 hz. Again, I assume that it is better to filter
while recording than after. How can I do this?

Use a scientific filter set for low pass, butterworth response, and about
tenth order.

Would filtering below 80 hz have much affect on a male
voice?

Depends on the voice and how it was miced. There may be some side effects
that are probably acceptable.

[Bill Bremmer[_2_]]

I've noted these readings with the mic switch off: At 4 o'clock -47 dB, at
12 o'clock -63, and at 9 o'clock -76 dB. As I said, I recorded a test with
the mic on and off at 4 o'clock. there is only a 2 or 3 dB difference when
the mic is on and when it's off. How do I know if the shield on the mic
cable is open or I have a ground loop on the system? The only things
connected to the PC are essentials, mouse, keyboard, monitor and the USB to
power the Mobile-Pre. No internet connection, printer etc. The switch that I
am using is the mic switch. I do not get any noise when I touch the mic or
plug and I don't move the mic.

I can try noise levels with the mic unplugged. Could this result be
different than just turning off the mic switch?

"Norbert Hahn" wrote in message
...
"Bill Bremmer" wrote:

I think the pre-amp cranked all the way up is producing the pre-amps own
system noise. Keep in mind that there is only about a 3dB difference on my
meter between the recorded voice and recording the noise with the mic
switch
off.

What kind of mic switch? A mute switch on the mic?
Anyway, your setup is broken. Either the shield of the mic cable is open,
or you have some ground loop in the system.
What external devices are connected to your PC?

What about using a condenser mike to get a stronger signal to noise ratio.
As I saidthe recorded signal only peaks at about 30%. Obviously I have to
normalize a lot to get a usable finished product.

A condensor mic may produce a 10 dB hotter signal - still not enough to
overcome a broken cable.

As you wrote that your recording suffer from some kind of static noise,
a simple low cut won't help you much. It will leave the higher frequencies
of the noise.

Do you get a clean recording when you set the gain to zero? Of course,
do don't record anything but white noise (hopefully).

What do you get recorded when you disconnect the mic from the mic amp,
with the gain setting to 4 o'clock?

Does the sound of the noise change when you touch the mic or the plug
or the cable?

Does the sound of the noise change when you move the mic around in the
room?

Norbert

[Arny Krueger]

"Bill Bremmer" wrote in message


I've noted these readings with the mic switch off: At 4
o'clock -47 dB, at 12 o'clock -63, and at 9 o'clock -76
dB. As I said, I recorded a test with the mic on and off
at 4 o'clock. there is only a 2 or 3 dB difference when
the mic is on and when it's off. How do I know if the
shield on the mic cable is open or I have a ground loop
on the system?

Mic on/off switches short the mic out in the "off" position.

If you've got a lot of hum with the mic switched "off" then you must have
some pretty serious problems elsewhere.

The most common problems with PC recording relate to grounding problems,
mostly ground loops.

Is the excess noise mostly composed of a hum? If you analyze it with Cool
Edit, is it mostly at the power line frequency and/or harmonics?

[Norbert Hahn]

"Bill Bremmer" wrote:

I've noted these readings with the mic switch off: At 4 o'clock -47 dB, at
12 o'clock -63, and at 9 o'clock -76 dB. As I said, I recorded a test with
the mic on and off at 4 o'clock. there is only a 2 or 3 dB difference when
the mic is on and when it's off.

That's strange. I don't know any mic amp that is as noisy as the
signal that comes from the mic.

How do I know if the shield on the mic
cable is open or I have a ground loop on the system? The only things
connected to the PC are essentials, mouse, keyboard, monitor and the USB to
power the Mobile-Pre. No internet connection, printer etc.

Ok, then there's no ground loop.

The switch that I
am using is the mic switch. I do not get any noise when I touch the mic or
plug and I don't move the mic.

Does the noise change when the mic is placed at a different location?
I suspect some stray magnetic fields that the mic picks up. Dynamic mic
may respond to magnetic fields.

I can try noise levels with the mic unplugged. Could this result be
different than just turning off the mic switch?

It would be strange if you get the same results.

BTW, I suffer from a similar problem, however, far less severe.
I have a new Edirol M-16DX which has 4 mic inputs and USB connection
to the computer. The inputs #1 and #4 generate a slight buzz, 60 dB
below maximum level on input #1 and 84 dB on input #4. The frequency
is 142 Hz, saw tooth! Nothing to heal with filtering because of the harmonics.
I'll sent the unit back for repair.

Norbert

[Bill Bremmer]

Paul and the group:

I made a new recording today, Monday August 30th.
With the system turned on and the mic powered but just sitting there in the
quiet room, here's what I noticed:

*With the gain set at 4 o'clock( the max is 6 o'clock); Mic switch on -45
dB; mic switch off -49 dB; mic unplugged -51 dB.

* With the gain set at 12 o'clock: mic switch on -64 dB; mic switch off -64
dB; mic unplugged -66 dB

* With the gain set at 9 o'clock: mic swich on -73 dB; mic switch off -73
dB; mic unplugged -75 dB

The recording is about 30 seconds long. I left about 4 seconds of space
before speaking and about four seconds of space after speaking. I did no
normalizing. In looking at the recording, I saw a noise level ("the space")
peaking at -44 dB. The voice level peaked at about -7 dB. The overall
recording was about 45% (of normalized?). The noise level looked to be
about 1%.

I talked to a tech at M Audio, the manufacturers of the Mobile-Pre. He
tested a unit there while I was on the phone. He said that he had similar
readings, about -51 dB at 4 o'clock with nothing plugged in. He said that I
need a condenser mic with phantom power to solve the problem.


"Paul Stamler" wrote in message
...
"Bill Bremmer" wrote in message
. ..
I'll try to keep one thread. But here is your answer. Cool Edit 2000 with
Studio Plug-In.
I've tried many of the filters supplied with Cool Edit (including making
notch filters of the noise frequencies) andhave had poor success. In the
instances the filter helped, the results were lost when I normalized.
Note:
there is only about a 3 dB difference showing on my meter between the
original voice recording and just recording with the mic switched off.

I think the pre-amp cranked all the way up is producing the pre-amps own
system noise. Keep in mind that there is only about a 3dB difference on
my
meter between the recorded voice and recording the noise with the mic
switch
off.

Well, now, you never mentioned *that*. Are you saying that the noise,
without normalizing, is only 3dB below the voice level? In which case it
must sound like you're talking at Niagara Falls. Or are you normalizing
the
voice signal and separately normalizing the noise? Because that would
invalidate every measurement you've mentioned.

What does the noise spectrum look like when the mic is switched off? Same
as
when it's on? Do the two measurements *without* normalizing; leave the
gain
set where you had it (4:00), do a short test with the mic on, another with
the mic off, and don't touch the normalizing function. Don't talk. What
are
the levels and frequencies?

What about using a condenser mike to get a stronger signal to noise
ratio.
As I saidthe recorded signal only peaks at about 30%. Obviously I have
to
normalize a lot to get a usable finished product.

You don't, really, but that's a very different issue. Let's leave it for
the
moment.

Peace,
Paul

[Scott Dorsey]

Bill Bremmer wrote:

I made a new recording today, Monday August 30th.
With the system turned on and the mic powered but just sitting there in the
quiet room, here's what I noticed:

*With the gain set at 4 o'clock( the max is 6 o'clock); Mic switch on -45
dB; mic switch off -49 dB; mic unplugged -51 dB.

* With the gain set at 12 o'clock: mic switch on -64 dB; mic switch off -64
dB; mic unplugged -66 dB

* With the gain set at 9 o'clock: mic swich on -73 dB; mic switch off -73
dB; mic unplugged -75 dB

What does it sound like with the mike shorted? Not unplugged, but shorted.

If you get similar characteristics with the mike shorted and unplugged, I
would characterize this as a noisy preamp.

I talked to a tech at M Audio, the manufacturers of the Mobile-Pre. He
tested a unit there while I was on the phone. He said that he had similar
readings, about -51 dB at 4 o'clock with nothing plugged in. He said that I
need a condenser mic with phantom power to solve the problem.

Or a preamp that isn't a piece of crap. You might look at the Lavry USB
interface instead. It has a preamp that isn't awful and is at least quiet
enough to use an RE-20 on, which for $250 is pretty damn good.
--scott
--
"C'est un Nagra. C'est suisse, et tres, tres precis."

[Paul Stamler]

"Bill Bremmer" wrote in message
. ..

I made a new recording today, Monday August 30th.
With the system turned on and the mic powered but just sitting there in
the
quiet room, here's what I noticed:

*With the gain set at 4 o'clock( the max is 6 o'clock); Mic switch on -45
dB; mic switch off -49 dB; mic unplugged -51 dB.

* With the gain set at 12 o'clock: mic switch on -64 dB; mic switch
off -64
dB; mic unplugged -66 dB

* With the gain set at 9 o'clock: mic swich on -73 dB; mic switch off -73
dB; mic unplugged -75 dB

The recording is about 30 seconds long. I left about 4 seconds of space
before speaking and about four seconds of space after speaking. I did no
normalizing. In looking at the recording, I saw a noise level ("the
space")
peaking at -44 dB. The voice level peaked at about -7 dB. The overall
recording was about 45% (of normalized?). The noise level looked to be
about 1%.

Okay, *now* we're cooking with gas. You can ignore the percentages; they're
just another way of describing the level. The 45% is percent of full-scale
(full-scale being digital saturation, or clipping), and it's the equivalent
of -6.9dB, the same as you saw on the dB scale.

Let's talk dB instead; it's a lot easier. With the gain set at 4:00, your
signal level is -7dBFS (that's "decibels relative to full-scale") and your
noise level is -45dBFS. That means your noise is only 38dB below your signal
level, and that's pretty bad.

So here's the next question: would it get better if you switched to a
higher-output microphone, like a condenser mic?

A clue to that is found within the first set of readings. When switched on,
you saw an increase in noise level, from -49dBFS to -45dBFS. That means the
microphone was adding some noise to the preamp's own noise. Some of that may
well have been the inherent resistive noise of the microphone's element, or
"Johnson noise"; if the off-switch shorts the microphone out then that's a
reasonable assumption. Some of it, however, may be room noise. Without
knowing a lot more about the inner workings of the preamp it's kind of hard
to know which.

However, the M-Audio guy's advice to try a condenser microphone, which will
have a higher output, now makes sense, given what we know about the test
results. So see if you can buy some sort of condenser mic from a place that
will let you return it and get a refund, perhaps minus a restocking fee, if
it doesn't help. Then try it out; set the gain to produce a similar -7dBFS
peak on the spoken words, and see what the noise looks like then.

If it's a lot better, then your problem was the preamp. If it's not, your
problem was the room noise. If it's only a little bit better, then it's
both.

In any case, at least we now have some useful numbers to work with --
thanks!

Peace,
Paul