Pages: 1 2 [3]   Go Down

Author Topic: Audio output DSPs and Parametric EQ speech enhancement  (Read 501 times)

KK4GMU

  • Member
  • Posts: 366
Re: Audio output DSPs and Parametric EQ speech enhancement
« Reply #30 on: July 19, 2022, 07:43:45 PM »

Your getting the cart before the horse.

Good Luck.
Mike K6AER
Mike, you are familiar with neither my horse nor my cart. My horse is lame; my cart has wobbly wheels. I love them both exactly as they are.  Nor with my neighborhood, my attic, my yard, my landscaping, my HOA, my budget, nor my intentions and modest aspirations in ham radio.  Not everyone wants to go 180 mph on a Yamaha.
Logged
IC-7100, RSPdx, AT-D878UVII-Plus HT, TGIF Spot

K7JQ

  • Member
  • Posts: 2602
Re: Audio output DSPs and Parametric EQ speech enhancement
« Reply #31 on: July 20, 2022, 08:07:54 AM »

Many of the *technical* posts here are well meaning, and trying to steer you to a more effective way of noise reduction before the speaker/headphone output. Yes, they are more effective, and require more extensive setup,  adjustment, and price. But they probably have never tried using a "digital magic box" plugged into a speaker/headphone jack, and can't believe they can be effective for your intended purpose.

No DSP NR circuit algorithm (outboard or built-in to a radio) is perfect, and they're not going to magically pull a weak, unintelligible signal out of the mud.  But the bottom line is they can take a signal to a more *comfortable* listening level by reducing the annoying underlying background noise/hiss from whatever source is causing it. Might not be totally effective on all sources. The intent is to maintain audio quality and volume with no digital artifacts as you adjust to decrease the background noise. Some units do it better than others.

The only way to find out if a particular unit is acceptable to you is to try it out and if not, sell it and try something else.

Bob K7JQ
Logged

KK4GMU

  • Member
  • Posts: 366
Re: Audio output DSPs and Parametric EQ speech enhancement
« Reply #32 on: July 20, 2022, 08:59:15 AM »

Many of the *technical* posts here are well meaning, and trying to steer you to a more effective way of noise reduction before the speaker/headphone output. Yes, they are more effective, and require more extensive setup,  adjustment, and price. But they probably have never tried using a "digital magic box" plugged into a speaker/headphone jack, and can't believe they can be effective for your intended purpose.

No DSP NR circuit algorithm (outboard or built-in to a radio) is perfect, and they're not going to magically pull a weak, unintelligible signal out of the mud.  But the bottom line is they can take a signal to a more *comfortable* listening level by reducing the annoying underlying background noise/hiss from whatever source is causing it. Might not be totally effective on all sources. The intent is to maintain audio quality and volume with no digital artifacts as you adjust to decrease the background noise. Some units do it better than others.

The only way to find out if a particular unit is acceptable to you is to try it out and if not, sell it and try something else.

Bob K7JQ
Good points.  I ordered the CLRdsp advertised on this site - shipped today.  Worth a shot.  Saved some $$ over new - hopefully.

Some well-meaning folks assume our objectives with ham radio are the same as theirs or are unaware of what those they attempt to assist have already tried, can afford, want to mess with, or want to accomplish.

In my case, I tried all the RF noise mitigation I care to try at this time.  I just want to try mitigation on the audio side.  Simple.  Trying my darndest to keep this thread on track:  Audio Noise Mitigation/Suppression to get the voice a bit more above the static on receive.
Logged
IC-7100, RSPdx, AT-D878UVII-Plus HT, TGIF Spot

K7JQ

  • Member
  • Posts: 2602
Re: Audio output DSPs and Parametric EQ speech enhancement
« Reply #33 on: July 20, 2022, 09:35:10 AM »

$140, no tax, shipped for a basically new WMR CLRdsp is a good deal (maybe you negotiated a lower price ;)). Push comes to shove, if you're not satisfied with it, you'll easily get your money back by selling it. Let us know how it works out for you.

Bob K7JQ
Logged

KK4GMU

  • Member
  • Posts: 366
Re: Audio output DSPs and Parametric EQ speech enhancement
« Reply #34 on: July 20, 2022, 10:17:52 AM »

$140, no tax, shipped for a basically new WMR CLRdsp is a good deal (maybe you negotiated a lower price ;)). Push comes to shove, if you're not satisfied with it, you'll easily get your money back by selling it. Let us know how it works out for you.

Bob K7JQ

I will.  No, I didn't negotiate.  I thought it was a fair deal already.  Thanks for bringing this to deal my attention.  I feel like you should get a buyers commission. 8)

My objective is to raise audible FM voice on 2 meters above the S9 static so I can better discern when its my turn to check in on our Marion County ERT net.  S9 is all around the exterior of the house per my HT walkabout.
That level extends beyond 100 feet around the perimeter to down the street a bit, on 2 meters. The source is not from within my house.
Logged
IC-7100, RSPdx, AT-D878UVII-Plus HT, TGIF Spot

AF5CC

  • Posts: 1664
    • HomeURL
Re: Audio output DSPs and Parametric EQ speech enhancement
« Reply #35 on: July 20, 2022, 01:12:37 PM »

Is it possible to put a beam up pointed at the repeater, or at the NCS station if on simplex?  That should help to increase the signal strength and also maybe null out some of the noise.  I know this is off topic from your thread, but it is another idea now that we know exactly what you are trying to accomplish.  2 meter beams are pretty small and it wouldn't need to be rotated if you just want to use it for the net.

73 John AF5CC
Logged

KK4GMU

  • Member
  • Posts: 366
Re: Audio output DSPs and Parametric EQ speech enhancement
« Reply #36 on: July 20, 2022, 05:53:35 PM »

Is it possible to put a beam up pointed at the repeater, or at the NCS station if on simplex?  That should help to increase the signal strength and also maybe null out some of the noise.  I know this is off topic from your thread, but it is another idea now that we know exactly what you are trying to accomplish.  2 meter beams are pretty small and it wouldn't need to be rotated if you just want to use it for the net.

73 John AF5CC

I thought briefly about that. But reaching the repeater with lower noise is just my immediate objective, not my only objective.  To lower the noise for signals from other sources and directions which I would also like to do could not be achieved with a unidirectional antenna without manually rotating it or installing a rotor $$.  Or getting another run of LMR-400 plus the cost of the antenna and another switch to go from the omni to the directional. Also, a beam, however small, could be just the ticket to get the attention of the HOA - moreso than my current stealthier vertical. 

There are a bunch of ways to address the RF options.  Another one, suggested by a local, is spending a few hundred bucks to achieve "best practices" optimum bonding and grounding using flat copper braided ground wire along with adding and interconnecting all my ground rods. Even that may or may not achieve any better results than what I already have.  https://www.gordonelectricsupply.com/p/Erico-557210-Flat-Cop-Braid/6479905?msclkid=84167602eefc151a0f2d14da4e2a1db3

To achieve my immediate and longer term objectives, trying the audio noise mitigation approach via a dsp unit is the better approach for me at this time. For my purposes, this is the better gamble. If this is unsatisfactory, Plans C, D, and E will be considered - potentially in my next life.  But radio might be obsolete by then (I'm looking ahead.) 

The CLRdsp has shipped.  It should be received and set up by next Tuesday.  I'll report results.
« Last Edit: July 20, 2022, 05:59:00 PM by KK4GMU »
Logged
IC-7100, RSPdx, AT-D878UVII-Plus HT, TGIF Spot

VK6HP

  • Member
  • Posts: 1222
Re: Audio output DSPs and Parametric EQ speech enhancement
« Reply #37 on: July 20, 2022, 09:29:35 PM »

You'll likely find that there is no silver bullet and that you end up having to approach various bands and modes differently.  That said, the threshold effect of even narrowband FM is just about the best thing you have going for you: although an inconvenient truth, it's a truth nonetheless that improving the signal-to-noise ratio prior to demodulation is a sound starting point and the gains from that will pay more than proportional s/n gains in the demodulated FM. So, the suggestion of a directional antenna to improve the s/n is a good one, particularly for FM.  If you were starting with a blank sheet (which presumably you're not) I'd also recommend a closer look at the FM receiver you're using: some have much better limiters and demodulators than others.

I use a CLRdsp unit, amongst other coherent and baseband processors in my station.  Leaving aside the much better returns from time invested in things like LF/MF/HF receive antennas, good common mode choking and (where possible) noise source suppression, I rate the CLRdsp unit as somewhat useful, without being outstanding.  The most successful application is with HF SSB using vintage receivers.  With no receiver DSP, my first and usually most successful weapon is to use what the receiver offers in the way of noise limiting (IF clipping), noise blanking (in newer classic receivers), and passband modification (via internal/external filters and/or passband shift/notch controls).  The baseband DSP processors sometimes, but not always, help somewhat beyond that.

When you get the CLRdsp give it a run over a day or two and see if you think it helps beyond some useful tonal modification.  I seem to remember that it uses a bridge audio output so follow the manufacturer's instructions regarding not grounding one side of the speaker; that was a minor hassle in my shack with multiple audio devices but it's easily addressed.

73, Peter.
Logged

AF5CC

  • Posts: 1664
    • HomeURL
Re: Audio output DSPs and Parametric EQ speech enhancement
« Reply #38 on: July 20, 2022, 09:53:37 PM »

You might need to see also if you can find a FM radio that has a noise blanker in it.  Very few if any amateur transceivers have noise blankers that function on FM, but I believe some commercial FM radios do, and most should operate in the 2 meter band with little to no modification.

73 John AF5CC
Logged

KK4GMU

  • Member
  • Posts: 366
Re: Audio output DSPs and Parametric EQ speech enhancement
« Reply #39 on: July 21, 2022, 07:15:40 AM »

My 7100 has a noise blanker, but not on 2 meters.  From my experience, NBs are effective for the right kind of noise - a fast beat-type or pulsing noise.

That is not the type of noise I have.  It is more of a smooth, consistent, staticy atmospheric noise.  I'm guessing a noise blanker would have minimal effect on this type of noise.  The NR on the other hand does have some modest to moderate effect, depending on how much I turn up the setting.

Again, I'm expecting one benefit of a CLRdsp-type device to be immediate accessibility to adjustment of the degree of noise suppression and tone on the fly rather than going into the 7100's menus.  Most radio NR adjustments suppress too much of the higher frequencies, dulling the the critical sibilants.  A concurrently adjustible tone control helps compensate for the loss of highs.
« Last Edit: July 21, 2022, 07:28:58 AM by KK4GMU »
Logged
IC-7100, RSPdx, AT-D878UVII-Plus HT, TGIF Spot

KK4GMU

  • Member
  • Posts: 366
Re: Audio output DSPs and Parametric EQ speech enhancement
« Reply #40 on: July 26, 2022, 04:43:50 AM »

You'll likely find that there is no silver bullet and that you end up having to approach various bands and modes differently.  That said, the threshold effect of even narrowband FM is just about the best thing you have going for you
True, that.  In fact, learning more about my 7100, I will do some reception testing with the FM bandwidth filter set to the narrowest which is 7K.  Using my SDR I discovered there is a narrow noise peak a MHz or two below my desired repeater RX frequency.  Not sure if that is a culprit or not.
Quote
: although an inconvenient truth, it's a truth nonetheless that improving the signal-to-noise ratio prior to demodulation is a sound starting point and the gains from that will pay more than proportional s/n gains in the demodulated FM. So, the suggestion of a directional antenna to improve the s/n is a good one, particularly for FM.  If you were starting with a blank sheet (which presumably you're not) I'd also recommend a closer look at the FM receiver you're using: some have much better limiters and demodulators than others.

I use a CLRdsp unit, amongst other coherent and baseband processors in my station.  Leaving aside the much better returns from time invested in things like LF/MF/HF receive antennas, good common mode choking and (where possible) noise source suppression, I rate the CLRdsp unit as somewhat useful, without being outstanding.  The most successful application is with HF SSB using vintage receivers.  With no receiver DSP, my first and usually most successful weapon is to use what the receiver offers in the way of noise limiting (IF clipping), noise blanking (in newer classic receivers), and passband modification (via internal/external filters and/or passband shift/notch controls).  The baseband DSP processors sometimes, but not always, help somewhat beyond that.

When you get the CLRdsp give it a run over a day or two and see if you think it helps beyond some useful tonal modification.  I seem to remember that it uses a bridge audio output so follow the manufacturer's instructions regarding not grounding one side of the speaker; that was a minor hassle in my shack with multiple audio devices but it's easily addressed.

73, Peter.
Thanks.  Yes, being one who likes to play with new stuff, I found a good reason to go to my local O'Reilly's Auto Parts store to buy a pack of five 2 amp fuses for $4.99.  In spite of the User Manual having a clear warning about "not grounding" and your statement above, of course I had to try to hook up my powered Klipsch speaker to the CLRdsp.  Two fuses later, I learned what works and what doesn't.  Some of us are either a) slower learners, or b) like to see if warnings are REALLY true.  I was happy that there was a fuse inside.

While my testing of the CLRdsp is not complete yet (I have a test session later this morning with another ham) here are my preliminary thoughts.
1. As expected, the immediate accessibility of the controls for degree of NR as well as the tone control (bringing up the highs suppressed by NR) is very useful and appreciated.
2. Regarding noise/static reduction on "mildly" staticy RX, there is a "slight" adavantage of the CLRdsp over the facility in the 7100. And my experience, despite some opinions to the contrary, is that a combination of the 7100 NR and the CLRdsp NR - in moderation - can result in an overall improvement.
3.  On severely staticy RX where voice is buried in the static such that Readability is 1 or 2, the jury is still out and subject to more testing.

I remain curious how the Xiegu GNR1 or the bhi dsp units stack up against the West Mountain CLRdsp.  I suspect they are within a gnats eyelash of one another (technical measurement).  Curious minds can be a PITA.

And thanks, Peter, 73s

Jerry
Logged
IC-7100, RSPdx, AT-D878UVII-Plus HT, TGIF Spot

KK4GMU

  • Member
  • Posts: 366
Re: Audio output DSPs and Parametric EQ speech enhancement
« Reply #41 on: July 28, 2022, 06:39:28 PM »

I participated on my Thursday nite net on the noisy repeater that I've had a 1 to 2 RX signal with in past weeks.

In combination of the IC-7100 IF filter being set to "3" (narrow - 7khz) and the CLRdsp filter set at 1 o'clock with "tone" full clockwise (treble), I had a solid 3.5+ on RX.

That repeater, frankly is the noisiest in the region, at least at my location. The several others I've used on 2 meter FM have RX at solid 5/9.  Hardly any hiss.

So I am learning that I have more of a single repeater RX noise problem than I have a local noise problem.

I am gradually entertaining the idea of a 5 element 2 meter yagi in my attic, especially after I thought through the logistics of installation.  Till now I thought I had to drag my 76 year old butt 15 feet through a foot of attic insulation but discovered a way to camouflage the cable within the house to the antenna mounting location next to the access hatch -  all with, the help of just a few feet of Wire Mold.
Logged
IC-7100, RSPdx, AT-D878UVII-Plus HT, TGIF Spot

KK4GMU

  • Member
  • Posts: 366
Re: Audio output DSPs and Parametric EQ speech enhancement
« Reply #42 on: August 05, 2022, 12:05:31 PM »

The noise on that one repeater frequency "miraculously" went away.  I had it for months. Last week it subsided somewhat.  Last night it was non-existent. My best guess is there was some relatively local noise source that manifest on that frequency that no longer exists.

Even so I am glad I have the DSP unit for immediately accessible noise reduction and tone control that works equal to or better than the NR in the radio.  Noise can be a recurring problem on any frequency.
Logged
IC-7100, RSPdx, AT-D878UVII-Plus HT, TGIF Spot
Pages: 1 2 [3]   Go Up