|
Sorry, another question... I was wondering what sort of processing the processor function on the transmit audio is. It seems to improve the average to peak power ratio but only by a relatively small amount. By the comment that it makes the audio harsh I rather assume it implements a similar function to rf clipping. I can’t find any description for it in the audio help. Am I right that the audio alc is syllabic in time constant? I have had a couple of reports that the audio is a bit thin which set me off on this. Playing with the audio filtering and the equaliser doesn’t quite get me what I want. The headset boom mic is not the most expensive I have to say. Mike Sent from Mail for Windows
|
|
|
Near any mic should give a goid audio... Equalizer plus processing... Maybe switch on the mic boost... Already cranked up mic audio in windows?? If you look in tx tab where the vox slider is.. While talking in the mic (no need to tx!).. How high does the grey bar come up?? (shows raw audio strength that goes IN the console)....
Processing with 30 to 40 should give enough boost...
Thin audio?? Boost low tones a bit..
Greetz sigi dg9bfc Am 27.08.2021 19:46 schrieb MikeC <mike.christieson@...>:
Sorry, another question... I was wondering what sort of processing the processor function on the transmit audio is. It seems to improve the average to peak power ratio but only by a relatively small amount. By the comment that it makes the audio harsh I rather assume it implements a similar function to rf clipping. I can’t find any description for it in the audio help. Am I right that the audio alc is syllabic in time constant? I have had a couple of reports that the audio is a bit thin which set me off on this. Playing with the audio filtering and the equaliser doesn’t quite get me what I want. The headset boom mic is not the most expensive I have to say. Mike Sent from Mail for Windows
|
|
|
Mike, The speech processor is quite good, it’s a compandor followed by a limiter. Let’s see a screenshot of the transmit options while you’re talking, the Spectrum shows the range and relative levels. A few tweaks and you’ll be good to go.
From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of MikeC
Sent: 27 August 2021 18:46
To: main@sdr-radio.groups.io
Subject: [SDR-Radio] Speech Processor Sorry, another question... I was wondering what sort of processing the processor function on the transmit audio is. It seems to improve the average to peak power ratio but only by a relatively small amount. By the comment that it makes the audio harsh I rather assume it implements a similar function to rf clipping. I can’t find any description for it in the audio help. Am I right that the audio alc is syllabic in time constant? I have had a couple of reports that the audio is a bit thin which set me off on this. Playing with the audio filtering and the equaliser doesn’t quite get me what I want. The headset boom mic is not the most expensive I have to say. Mike Sent from Mail for Windows -- - + - + -
|
|
|
kb3cs
|
|
|
Probably. I use a single band RMS compressor followed by a very-well designed limiter which avoids overshoot. I’ve tried multi-band compressors but not achieved much benefit. When I get into my main winter project I may look at this again, having a compressor instance for each octave or half-octave. For winter I want to finally play with my Hermes-Lite 2, with my main amplifier I’ll have ~350 watts so will want to boost the audio. Simon Brown, G4ELI https://www.sdr-radio.com
|
|
|
Interesting article. I did experiment further with the processor settings and compared the result with an analogue speech processor that was part of a ssb rig I designed about 20 years ago. This processor uses variable rf log clipping of an ssb signal (at 10.7MHz) followed by another ssb filter. The audio is pre-processed by a syllabic compressor (a vogad). The clipping is done with two stages of log amplifiers. In the original design the resulting ssb was converted to the required frequency but I had already added a demodulator back down to audio for something else so here the output is audio. As the article points outs, non dc baseband clipping avoids the serious harmonic distortion that baseband clipping induces at all but very light clipping because the harmonics multiply outside the passband. There are still intermodulation products that eventually limit performance. Of course this processor is very component hungry by today’s standards and I happened to have it on the shelf. As the article also points out, the the same result can be achieved in dsp by other means that could not be realised in practice by analogue circuits. I feed the audio output at line level into the pc and have the SDRC processor turned off. The audio processor bandwidth is from about 200Hz to 2.4kHz, determined almost entirely by the analogue ssb filters. I think that overshoot is a slightly different, but related problem. Interestingly though I had a look at this processor’s performance by putting test sine and square-waves frequencies into it and looking at the audio output on a scope. Probably by more luck than judgement, the overshoot is a few percent unless the processing is set ridiculously high. I suspect that a fortuitous combination of filters and time constants, plus the relatively narrow bandwidth requirement helps. I have not checked for overshoot on the rf output from the Limesdr that eventually generates the signal for transmission. My metering indicates that the average to peak power ratio has increased significantly and, although this is pretty unscientific, the “hf pileup breaking index” is markedly improved. Nothing can be all things to all men (or women). Everything is a compromise of requirements, performance, complexity, time and cost. Mike Sent from Mail for Windows
From: Simon BrownSent: 31 August 2021 06:41 To: main@SDR-Radio.groups.ioSubject: Re: [SDR-Radio] Speech Processor Probably. I use a single band RMS compressor followed by a very-well designed limiter which avoids overshoot. I’ve tried multi-band compressors but not achieved much benefit. When I get into my main winter project I may look at this again, having a compressor instance for each octave or half-octave. For winter I want to finally play with my Hermes-Lite 2, with my main amplifier I’ll have ~350 watts so will want to boost the audio. Simon Brown, G4ELI https://www.sdr-radio.com From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of kb3cs
Sent: 30 August 2021 21:37
To: main@SDR-Radio.groups.io
Subject: Re: [SDR-Radio] Speech Processor Controlled-Envelope good?
http://www.arrl.org/files/file/QEX_Next_Issue/2014/Nov-Dec_2014/Hershberger_QEX_11_14.pdf
- 73 -
--
|
|
|
the nice thing when using all in software (sdr) you can do things
with compression, limiter, etc etc ... that you can NOT do with
analogue circuits
your "rf clipper" may produce a more boosted signal ... but i
guess when all is made in software you can be a bit better
the sdr has the audio in a chain (buffer) and you can have a look
forward agc ... limit every word to the maximum without
overshooting or harmonics produced
and with a monitor receiver on the output of your amp you can add
predistortion ... result is an even cleaner (but more punchy)
signal
no way to do that with conventional circuits
you could try to combine your rf clipper with the processor in
sdrc (but do not crank it up tooo much) ... i do a simlar thingy
with my mixing software (voicemeeter potatoe) with some vst
plugins added in the chain ... first the plugins do a tiny bit of
compression and limits the peaks not to overdrive the sdrc input
... and a bit of processing of the sdrc is added afterwards ...
the result is a well boosted signal with no splatter (pay also
attention not to produce intermod distortion on the final amp or
driving stages!!)
simon you have the equalizer ... what if each filter in the eq
would have its own processor and limiter?? is that what you mean
with half octave filters (and boosting those) ?!?!?
dg9bfc sigi
Am 31.08.2021 um 12:01 schrieb MikeC:
toggle quoted messageShow quoted text
Interesting article. I did experiment
further with the processor settings and compared the result
with an analogue speech processor that was part of a ssb rig I
designed about 20 years ago. This processor uses variable rf
log clipping of an ssb signal (at 10.7MHz) followed by another
ssb filter. The audio is pre-processed by a syllabic
compressor (a vogad). The clipping is done with two stages of
log amplifiers. In the original design the resulting ssb was
converted to the required frequency but I had already added a
demodulator back down to audio for something else so here the
output is audio. As the article points outs, non dc baseband
clipping avoids the serious harmonic distortion that baseband
clipping induces at all but very light clipping because the
harmonics multiply outside the passband. There are still
intermodulation products that eventually limit performance.
Of course this processor is very component
hungry by today’s standards and I happened to have it on the
shelf. As the article also points out, the the same result can
be achieved in dsp by other means that could not be realised
in practice by analogue circuits.
I feed the audio output at line level into
the pc and have the SDRC processor turned off. The audio
processor bandwidth is from about 200Hz to 2.4kHz, determined
almost entirely by the analogue ssb filters.
I think that overshoot is a slightly
different, but related problem. Interestingly though I had a
look at this processor’s performance by putting test sine and
square-waves frequencies into it and looking at the audio
output on a scope. Probably by more luck than judgement, the
overshoot is a few percent unless the processing is set
ridiculously high. I suspect that a fortuitous combination of
filters and time constants, plus the relatively narrow
bandwidth requirement helps.
I have not checked for overshoot on the rf
output from the Limesdr that eventually generates the signal
for transmission.
My metering indicates that the average to
peak power ratio has increased significantly and, although
this is pretty unscientific, the “hf pileup breaking index” is
markedly improved.
Nothing can be all things to all men (or
women). Everything is a compromise of requirements,
performance, complexity, time and cost.
Mike
Sent from Mail for Windows
Probably.
I
use a single band RMS compressor followed by a very-well
designed limiter which avoids overshoot. I’ve tried
multi-band compressors but not achieved much benefit. When I
get into my main winter project I may look at this again,
having a compressor instance for each octave or half-octave.
For
winter I want to finally play with my Hermes-Lite 2, with my
main amplifier I’ll have ~350 watts so will want to boost
the audio.
Simon Brown, G4ELI
https://www.sdr-radio.com
Controlled-Envelope good?
http://www.arrl.org/files/file/QEX_Next_Issue/2014/Nov-Dec_2014/Hershberger_QEX_11_14.pdf
- 73 -
--
|
|
|
Sigi as you are using VST plugins have you tried the Waves L3 Multimaximizer? It is hard to make it sound bad and it certainly increases talk power with minimal distortion, it never overshoots.
73
Conrad
toggle quoted messageShow quoted text
From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io>
On Behalf Of Siegfried Jackstien via groups.io
Sent: 31 August 2021 12:22
To: main@SDR-Radio.groups.io
Subject: Re: [SDR-Radio] Speech Processor
the nice thing when using all in software (sdr) you can do things with compression, limiter, etc etc ... that you can NOT do with analogue circuits
your "rf clipper" may produce a more boosted signal ... but i guess when all is made in software you can be a bit better
the sdr has the audio in a chain (buffer) and you can have a look forward agc ... limit every word to the maximum without overshooting or harmonics produced
and with a monitor receiver on the output of your amp you can add predistortion ... result is an even cleaner (but more punchy) signal
no way to do that with conventional circuits
you could try to combine your rf clipper with the processor in sdrc (but do not crank it up tooo much) ... i do a simlar thingy with my mixing software (voicemeeter potatoe) with some vst plugins added in the chain ... first the plugins do a tiny bit of
compression and limits the peaks not to overdrive the sdrc input ... and a bit of processing of the sdrc is added afterwards ... the result is a well boosted signal with no splatter (pay also attention not to produce intermod distortion on the final amp or
driving stages!!)
simon you have the equalizer ... what if each filter in the eq would have its own processor and limiter?? is that what you mean with half octave filters (and boosting those) ?!?!?
dg9bfc sigi
Am 31.08.2021 um 12:01 schrieb MikeC:
Interesting article. I did experiment further with the processor settings and compared the result with an analogue speech processor that was part of a ssb rig I designed about 20 years ago. This processor uses variable rf log clipping
of an ssb signal (at 10.7MHz) followed by another ssb filter. The audio is pre-processed by a syllabic compressor (a vogad). The clipping is done with two stages of log amplifiers. In the original design the resulting ssb was converted to the required frequency
but I had already added a demodulator back down to audio for something else so here the output is audio. As the article points outs, non dc baseband clipping avoids the serious harmonic distortion that baseband clipping induces at all but very light clipping
because the harmonics multiply outside the passband. There are still intermodulation products that eventually limit performance.
Of course this processor is very component hungry by today’s standards and I happened to have it on the shelf. As the article also points out, the the same result can be achieved in dsp by other means that could not be realised in practice
by analogue circuits.
I feed the audio output at line level into the pc and have the SDRC processor turned off. The audio processor bandwidth is from about 200Hz to 2.4kHz, determined almost entirely by the analogue ssb filters.
I think that overshoot is a slightly different, but related problem. Interestingly though I had a look at this processor’s performance by putting test sine and square-waves frequencies into it and looking at the audio output on a scope.
Probably by more luck than judgement, the overshoot is a few percent unless the processing is set ridiculously high. I suspect that a fortuitous combination of filters and time constants, plus the relatively narrow bandwidth requirement helps.
I have not checked for overshoot on the rf output from the Limesdr that eventually generates the signal for transmission.
My metering indicates that the average to peak power ratio has increased significantly and, although this is pretty unscientific, the “hf pileup breaking index” is markedly improved.
Nothing can be all things to all men (or women). Everything is a compromise of requirements, performance, complexity, time and cost.
Mike
Sent from
Mail for Windows
Probably.
I use a single band RMS compressor followed by a very-well designed limiter which avoids overshoot. I’ve tried multi-band compressors but not achieved much benefit. When I get into my main winter project I may look at this again, having
a compressor instance for each octave or half-octave.
For winter I want to finally play with my Hermes-Lite 2, with my main amplifier I’ll have ~350 watts so will want to boost the audio.
Simon Brown, G4ELI
https://www.sdr-radio.com
Controlled-Envelope good?
http://www.arrl.org/files/file/QEX_Next_Issue/2014/Nov-Dec_2014/Hershberger_QEX_11_14.pdf
- 73 -
--
|
|
|
One thing, With software we can’t have overshoot, hence a well-designed limiter. In software it’s also easy to write an RF speech processor but when set too high can sound harsh. I get favourable reports on my audio with Pluto and FDM-DUO so am happy. Now to get my HL2 on the air…
From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of MikeC
Sent: 31 August 2021 11:01
To: main@SDR-Radio.groups.io
Subject: Re: [SDR-Radio] Speech Processor Interesting article. I did experiment further with the processor settings and compared the result with an analogue speech processor that was part of a ssb rig I designed about 20 years ago. This processor uses variable rf log clipping of an ssb signal (at 10.7MHz) followed by another ssb filter. The audio is pre-processed by a syllabic compressor (a vogad). The clipping is done with two stages of log amplifiers. In the original design the resulting ssb was converted to the required frequency but I had already added a demodulator back down to audio for something else so here the output is audio. As the article points outs, non dc baseband clipping avoids the serious harmonic distortion that baseband clipping induces at all but very light clipping because the harmonics multiply outside the passband. There are still intermodulation products that eventually limit performance. Of course this processor is very component hungry by today’s standards and I happened to have it on the shelf. As the article also points out, the the same result can be achieved in dsp by other means that could not be realised in practice by analogue circuits. I feed the audio output at line level into the pc and have the SDRC processor turned off. The audio processor bandwidth is from about 200Hz to 2.4kHz, determined almost entirely by the analogue ssb filters. I think that overshoot is a slightly different, but related problem. Interestingly though I had a look at this processor’s performance by putting test sine and square-waves frequencies into it and looking at the audio output on a scope. Probably by more luck than judgement, the overshoot is a few percent unless the processing is set ridiculously high. I suspect that a fortuitous combination of filters and time constants, plus the relatively narrow bandwidth requirement helps. I have not checked for overshoot on the rf output from the Limesdr that eventually generates the signal for transmission. My metering indicates that the average to peak power ratio has increased significantly and, although this is pretty unscientific, the “hf pileup breaking index” is markedly improved. Nothing can be all things to all men (or women). Everything is a compromise of requirements, performance, complexity, time and cost. Mike Sent from Mail for Windows Probably. I use a single band RMS compressor followed by a very-well designed limiter which avoids overshoot. I’ve tried multi-band compressors but not achieved much benefit. When I get into my main winter project I may look at this again, having a compressor instance for each octave or half-octave. For winter I want to finally play with my Hermes-Lite 2, with my main amplifier I’ll have ~350 watts so will want to boost the audio. Simon Brown, G4ELI https://www.sdr-radio.com Controlled-Envelope good?
http://www.arrl.org/files/file/QEX_Next_Issue/2014/Nov-Dec_2014/Hershberger_QEX_11_14.pdf
- 73 -
--
-- - + - + -
|
|
|
i do not know it ... (the l3 multimaximizer)
... (one or two minutes later) ... ok found it ...
lots of free vst plugins available so i will definitely not PAY
40 bucks for a plugin ;-)
my used plugins (in order how audio runs) ...
reafir (surpress some noise ... especially some "rumble" on the
low end)
reaeq (shape the input signal a bit)
reagate (only send audio when you are speaking but low
background noise is surpressed)
reaxcomp (compressor with multiple bands) ... seems to be similar
to the l3
protoverb (just slight delay to give the audio a more filled
sound)
and finally frontier (self adaptive limiter) ... that combination
wortks superb BUT:
such a high number of plugins needs a lot of tweaking till your
audio sounds like from the bbc newsreader (grin) so do not ask me
how i have set it all ... it would take a much longer mail to
describe what i have done here
so finetune your audio (or making a complete mess from it) can be
done easy hi hi
the l3 multimaximizer seems to be similar to reaxcomp so i do not
need to try out the L3 i guess
greetz sigi dg9bfc
Am 31.08.2021 um 13:43 schrieb Conrad,
PA5Y:
toggle quoted messageShow quoted text
Sigi as you are using VST plugins have you
tried the Waves L3 Multimaximizer? It is hard to make it sound
bad and it certainly increases talk power with minimal
distortion, it never overshoots.
73
Conrad
the nice thing when using all in software (sdr) you can do
things with compression, limiter, etc etc ... that you can NOT
do with analogue circuits
your "rf clipper" may produce a more boosted signal ... but i
guess when all is made in software you can be a bit better
the sdr has the audio in a chain (buffer) and you can have a
look forward agc ... limit every word to the maximum without
overshooting or harmonics produced
and with a monitor receiver on the output of your amp you can
add predistortion ... result is an even cleaner (but more
punchy) signal
no way to do that with conventional circuits
you could try to combine your rf clipper with the processor
in sdrc (but do not crank it up tooo much) ... i do a simlar
thingy with my mixing software (voicemeeter potatoe) with some
vst plugins added in the chain ... first the plugins do a tiny
bit of compression and limits the peaks not to overdrive the
sdrc input ... and a bit of processing of the sdrc is added
afterwards ... the result is a well boosted signal with no
splatter (pay also attention not to produce intermod
distortion on the final amp or driving stages!!)
simon you have the equalizer ... what if each filter in the
eq would have its own processor and limiter?? is that what you
mean with half octave filters (and boosting those) ?!?!?
dg9bfc sigi
Am 31.08.2021 um 12:01 schrieb MikeC:
Interesting article. I did experiment
further with the processor settings and compared the result
with an analogue speech processor that was part of a ssb rig
I designed about 20 years ago. This processor uses variable
rf log clipping of an ssb signal (at 10.7MHz) followed by
another ssb filter. The audio is pre-processed by a syllabic
compressor (a vogad). The clipping is done with two stages
of log amplifiers. In the original design the resulting ssb
was converted to the required frequency but I had already
added a demodulator back down to audio for something else so
here the output is audio. As the article points outs, non dc
baseband clipping avoids the serious harmonic distortion
that baseband clipping induces at all but very light
clipping because the harmonics multiply outside the
passband. There are still intermodulation products that
eventually limit performance.
Of course this processor is very
component hungry by today’s standards and I happened to have
it on the shelf. As the article also points out, the the
same result can be achieved in dsp by other means that could
not be realised in practice by analogue circuits.
I feed the audio output at line level
into the pc and have the SDRC processor turned off. The
audio processor bandwidth is from about 200Hz to 2.4kHz,
determined almost entirely by the analogue ssb filters.
I think that overshoot is a slightly
different, but related problem. Interestingly though I had a
look at this processor’s performance by putting test sine
and square-waves frequencies into it and looking at the
audio output on a scope. Probably by more luck than
judgement, the overshoot is a few percent unless the
processing is set ridiculously high. I suspect that a
fortuitous combination of filters and time constants, plus
the relatively narrow bandwidth requirement helps.
I have not checked for overshoot on the
rf output from the Limesdr that eventually generates the
signal for transmission.
My metering indicates that the average to
peak power ratio has increased significantly and, although
this is pretty unscientific, the “hf pileup breaking index”
is markedly improved.
Nothing can be all things to all men (or
women). Everything is a compromise of requirements,
performance, complexity, time and cost.
Mike
Sent from
Mail for Windows
Probably.
I use a single band RMS compressor
followed by a very-well designed limiter which avoids
overshoot. I’ve tried multi-band compressors but not
achieved much benefit. When I get into my main winter
project I may look at this again, having a compressor
instance for each octave or half-octave.
For winter I want to finally play with my
Hermes-Lite 2, with my main amplifier I’ll have ~350 watts
so will want to boost the audio.
Simon Brown, G4ELI
https://www.sdr-radio.com
Controlled-Envelope good?
http://www.arrl.org/files/file/QEX_Next_Issue/2014/Nov-Dec_2014/Hershberger_QEX_11_14.pdf
- 73 -
--
|
|
|
Or, Just buy a decent microphone?
From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of Siegfried Jackstien
Sent: 31 August 2021 16:23
To: main@SDR-Radio.groups.io
Subject: Re: [SDR-Radio] Speech Processor i do not know it ... (the l3 multimaximizer) ... (one or two minutes later) ... ok found it ... lots of free vst plugins available so i will definitely not PAY 40 bucks for a plugin ;-) my used plugins (in order how audio runs) ... reafir (surpress some noise ... especially some "rumble" on the low end) reaeq (shape the input signal a bit) reagate (only send audio when you are speaking but low background noise is surpressed) reaxcomp (compressor with multiple bands) ... seems to be similar to the l3 protoverb (just slight delay to give the audio a more filled sound) and finally frontier (self adaptive limiter) ... that combination wortks superb BUT: such a high number of plugins needs a lot of tweaking till your audio sounds like from the bbc newsreader (grin) so do not ask me how i have set it all ... it would take a much longer mail to describe what i have done here so finetune your audio (or making a complete mess from it) can be done easy hi hi the l3 multimaximizer seems to be similar to reaxcomp so i do not need to try out the L3 i guess greetz sigi dg9bfc Am 31.08.2021 um 13:43 schrieb Conrad, PA5Y: Sigi as you are using VST plugins have you tried the Waves L3 Multimaximizer? It is hard to make it sound bad and it certainly increases talk power with minimal distortion, it never overshoots. 73 Conrad the nice thing when using all in software (sdr) you can do things with compression, limiter, etc etc ... that you can NOT do with analogue circuits your "rf clipper" may produce a more boosted signal ... but i guess when all is made in software you can be a bit better the sdr has the audio in a chain (buffer) and you can have a look forward agc ... limit every word to the maximum without overshooting or harmonics produced and with a monitor receiver on the output of your amp you can add predistortion ... result is an even cleaner (but more punchy) signal no way to do that with conventional circuits you could try to combine your rf clipper with the processor in sdrc (but do not crank it up tooo much) ... i do a simlar thingy with my mixing software (voicemeeter potatoe) with some vst plugins added in the chain ... first the plugins do a tiny bit of compression and limits the peaks not to overdrive the sdrc input ... and a bit of processing of the sdrc is added afterwards ... the result is a well boosted signal with no splatter (pay also attention not to produce intermod distortion on the final amp or driving stages!!) simon you have the equalizer ... what if each filter in the eq would have its own processor and limiter?? is that what you mean with half octave filters (and boosting those) ?!?!? dg9bfc sigi Am 31.08.2021 um 12:01 schrieb MikeC: Interesting article. I did experiment further with the processor settings and compared the result with an analogue speech processor that was part of a ssb rig I designed about 20 years ago. This processor uses variable rf log clipping of an ssb signal (at 10.7MHz) followed by another ssb filter. The audio is pre-processed by a syllabic compressor (a vogad). The clipping is done with two stages of log amplifiers. In the original design the resulting ssb was converted to the required frequency but I had already added a demodulator back down to audio for something else so here the output is audio. As the article points outs, non dc baseband clipping avoids the serious harmonic distortion that baseband clipping induces at all but very light clipping because the harmonics multiply outside the passband. There are still intermodulation products that eventually limit performance. Of course this processor is very component hungry by today’s standards and I happened to have it on the shelf. As the article also points out, the the same result can be achieved in dsp by other means that could not be realised in practice by analogue circuits. I feed the audio output at line level into the pc and have the SDRC processor turned off. The audio processor bandwidth is from about 200Hz to 2.4kHz, determined almost entirely by the analogue ssb filters. I think that overshoot is a slightly different, but related problem. Interestingly though I had a look at this processor’s performance by putting test sine and square-waves frequencies into it and looking at the audio output on a scope. Probably by more luck than judgement, the overshoot is a few percent unless the processing is set ridiculously high. I suspect that a fortuitous combination of filters and time constants, plus the relatively narrow bandwidth requirement helps. I have not checked for overshoot on the rf output from the Limesdr that eventually generates the signal for transmission. My metering indicates that the average to peak power ratio has increased significantly and, although this is pretty unscientific, the “hf pileup breaking index” is markedly improved. Nothing can be all things to all men (or women). Everything is a compromise of requirements, performance, complexity, time and cost. Mike Sent from Mail for Windows Probably. I use a single band RMS compressor followed by a very-well designed limiter which avoids overshoot. I’ve tried multi-band compressors but not achieved much benefit. When I get into my main winter project I may look at this again, having a compressor instance for each octave or half-octave. For winter I want to finally play with my Hermes-Lite 2, with my main amplifier I’ll have ~350 watts so will want to boost the audio. Simon Brown, G4ELI https://www.sdr-radio.com Controlled-Envelope good?
http://www.arrl.org/files/file/QEX_Next_Issue/2014/Nov-Dec_2014/Hershberger_QEX_11_14.pdf
- 73 -
--
-- - + - + -
|
|
|
Or take singing lessons....
Gedas, W8BYA EN70JT
Gallery at http://w8bya.com (under repair)
Light travels faster than sound....
This is why some people appear bright until you hear them speak.
On 8/31/2021 11:45 AM, Simon Brown
wrote:
toggle quoted messageShow quoted text
Or,
Just
buy a decent microphone?
i do not know it ... (the l3 multimaximizer)
... (one or two minutes later) ... ok found it ...
lots of free vst plugins available so i will definitely not
PAY 40 bucks for a plugin ;-)
my used plugins (in order how audio runs) ...
reafir (surpress some noise ... especially some "rumble" on
the low end)
reaeq (shape the input signal a bit)
reagate (only send audio when you are speaking but low
background noise is surpressed)
reaxcomp (compressor with multiple bands) ... seems to be
similar to the l3
protoverb (just slight delay to give the audio a more filled
sound)
and finally frontier (self adaptive limiter) ... that
combination wortks superb BUT:
such a high number of plugins needs a lot of tweaking till
your audio sounds like from the bbc newsreader (grin) so do
not ask me how i have set it all ... it would take a much
longer mail to describe what i have done here
so finetune your audio (or making a complete mess from it)
can be done easy hi hi
the l3 multimaximizer seems to be similar to reaxcomp so i do
not need to try out the L3 i guess
greetz sigi dg9bfc
Am 31.08.2021 um 13:43 schrieb Conrad,
PA5Y:
Sigi as you are using VST plugins have
you tried the Waves L3 Multimaximizer? It is hard to make it
sound bad and it certainly increases talk power with minimal
distortion, it never overshoots.
73
Conrad
the nice thing when using all in software (sdr) you can do
things with compression, limiter, etc etc ... that you can
NOT do with analogue circuits
your "rf clipper" may produce a more boosted signal ... but
i guess when all is made in software you can be a bit better
the sdr has the audio in a chain (buffer) and you can have
a look forward agc ... limit every word to the maximum
without overshooting or harmonics produced
and with a monitor receiver on the output of your amp you
can add predistortion ... result is an even cleaner (but
more punchy) signal
no way to do that with conventional circuits
you could try to combine your rf clipper with the processor
in sdrc (but do not crank it up tooo much) ... i do a simlar
thingy with my mixing software (voicemeeter potatoe) with
some vst plugins added in the chain ... first the plugins do
a tiny bit of compression and limits the peaks not to
overdrive the sdrc input ... and a bit of processing of the
sdrc is added afterwards ... the result is a well boosted
signal with no splatter (pay also attention not to produce
intermod distortion on the final amp or driving stages!!)
simon you have the equalizer ... what if each filter in the
eq would have its own processor and limiter?? is that what
you mean with half octave filters (and boosting those) ?!?!?
dg9bfc sigi
Am 31.08.2021 um 12:01 schrieb MikeC:
Interesting article. I did experiment
further with the processor settings and compared the
result with an analogue speech processor that was part of
a ssb rig I designed about 20 years ago. This processor
uses variable rf log clipping of an ssb signal (at
10.7MHz) followed by another ssb filter. The audio is
pre-processed by a syllabic compressor (a vogad). The
clipping is done with two stages of log amplifiers. In the
original design the resulting ssb was converted to the
required frequency but I had already added a demodulator
back down to audio for something else so here the output
is audio. As the article points outs, non dc baseband
clipping avoids the serious harmonic distortion that
baseband clipping induces at all but very light clipping
because the harmonics multiply outside the passband. There
are still intermodulation products that eventually limit
performance.
Of course this processor is very
component hungry by today’s standards and I happened to
have it on the shelf. As the article also points out, the
the same result can be achieved in dsp by other means that
could not be realised in practice by analogue circuits.
I feed the audio output at line level
into the pc and have the SDRC processor turned off. The
audio processor bandwidth is from about 200Hz to 2.4kHz,
determined almost entirely by the analogue ssb filters.
I think that overshoot is a slightly
different, but related problem. Interestingly though I had
a look at this processor’s performance by putting test
sine and square-waves frequencies into it and looking at
the audio output on a scope. Probably by more luck than
judgement, the overshoot is a few percent unless the
processing is set ridiculously high. I suspect that a
fortuitous combination of filters and time constants, plus
the relatively narrow bandwidth requirement helps.
I have not checked for overshoot on the
rf output from the Limesdr that eventually generates the
signal for transmission.
My metering indicates that the average
to peak power ratio has increased significantly and,
although this is pretty unscientific, the “hf pileup
breaking index” is markedly improved.
Nothing can be all things to all men
(or women). Everything is a compromise of requirements,
performance, complexity, time and cost.
Mike
Sent from Mail for Windows
Probably.
I use a single band RMS compressor
followed by a very-well designed limiter which avoids
overshoot. I’ve tried multi-band compressors but not
achieved much benefit. When I get into my main winter
project I may look at this again, having a compressor
instance for each octave or half-octave.
For winter I want to finally play with
my Hermes-Lite 2, with my main amplifier I’ll have ~350
watts so will want to boost the audio.
Simon Brown, G4ELI
https://www.sdr-radio.com
Controlled-Envelope good?
http://www.arrl.org/files/file/QEX_Next_Issue/2014/Nov-Dec_2014/Hershberger_QEX_11_14.pdf
- 73 -
--
--
- + - + -
|
|
|
I feel very disadvantaged ONLY having about a dozen old analogue chips in my processor...... Of course if I did it again now it would be dsp which would give much more flexibility as well. I do think the key is avoiding the harmonic distortion problem which dsp can do. In the end its all the same signal processing methods implemented by either an analogue or a digital computer. An old friend of mine (long dead now) who was a tutor at Imperial College said, “the interesting problems getting digital systems to work are actually analogue”. He was a control man and was also heard to say “any frequency above 1Hz is dangerously high”. Mike Sent from Mail for Windows
From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of Siegfried Jackstien
Sent: 31 August 2021 16:23
To: main@SDR-Radio.groups.io
Subject: Re: [SDR-Radio] Speech Processor i do not know it ... (the l3 multimaximizer) ... (one or two minutes later) ... ok found it ... lots of free vst plugins available so i will definitely not PAY 40 bucks for a plugin ;-) my used plugins (in order how audio runs) ... reafir (surpress some noise ... especially some "rumble" on the low end) reaeq (shape the input signal a bit) reagate (only send audio when you are speaking but low background noise is surpressed) reaxcomp (compressor with multiple bands) ... seems to be similar to the l3 protoverb (just slight delay to give the audio a more filled sound) and finally frontier (self adaptive limiter) ... that combination wortks superb BUT: such a high number of plugins needs a lot of tweaking till your audio sounds like from the bbc newsreader (grin) so do not ask me how i have set it all ... it would take a much longer mail to describe what i have done here so finetune your audio (or making a complete mess from it) can be done easy hi hi the l3 multimaximizer seems to be similar to reaxcomp so i do not need to try out the L3 i guess greetz sigi dg9bfc
|
|
|
I have a working DATONG speech clipper. Over 40 years old, sitting in a cupboard…
From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of MikeC
Sent: 31 August 2021 17:32
To: main@SDR-Radio.groups.io
Subject: Re: [SDR-Radio] Speech Processor I feel very disadvantaged ONLY having about a dozen old analogue chips in my processor...... Of course if I did it again now it would be dsp which would give much more flexibility as well. I do think the key is avoiding the harmonic distortion problem which dsp can do. In the end its all the same signal processing methods implemented by either an analogue or a digital computer. An old friend of mine (long dead now) who was a tutor at Imperial College said, “the interesting problems getting digital systems to work are actually analogue”. He was a control man and was also heard to say “any frequency above 1Hz is dangerously high”. Mike Sent from Mail for Windows i do not know it ... (the l3 multimaximizer) ... (one or two minutes later) ... ok found it ... lots of free vst plugins available so i will definitely not PAY 40 bucks for a plugin ;-) my used plugins (in order how audio runs) ... reafir (surpress some noise ... especially some "rumble" on the low end) reaeq (shape the input signal a bit) reagate (only send audio when you are speaking but low background noise is surpressed) reaxcomp (compressor with multiple bands) ... seems to be similar to the l3 protoverb (just slight delay to give the audio a more filled sound) and finally frontier (self adaptive limiter) ... that combination wortks superb BUT: such a high number of plugins needs a lot of tweaking till your audio sounds like from the bbc newsreader (grin) so do not ask me how i have set it all ... it would take a much longer mail to describe what i have done here so finetune your audio (or making a complete mess from it) can be done easy hi hi the l3 multimaximizer seems to be similar to reaxcomp so i do not need to try out the L3 i guess greetz sigi dg9bfc -- - + - + -
|
|
|
On 8/31/21 12:37 PM, Simon Brown wrote: I have a working DATONG speech clipper. Over 40 years old, sitting in a cupboard… That's fine as long as there are also tea and biscuits.
|
|
|
hmm i have a good microphone ... wireless speakermike with ptt
... or in other words a mic with a 600 feet long cable hi hi
barefoot it sound "ok" (with some equalizer tweaking and a bit of
compression added) ... but with the vst plugins it is a bit more
punchy (without sounding bad)
mainly i installed the mixer cause i wanted to have a few
different audio ways for rx in parallel (drive speaker, headphones
and wireless mike plus maybe the bluetooth speakers or the
speakers fron the tv set that i use as big monitor for sdrc)
the mixer on the tx part is also nice (easy to change levels also
for digimodes ...or whatever) ... the vst plugins just added as
"nice to have but not really needed"
no i will not take singing lessons (grin)
greetz sigi dg9bfc
Am 31.08.2021 um 17:45 schrieb Simon
Brown:
toggle quoted messageShow quoted text
Or,
Just
buy a decent microphone?
i do not know it ... (the l3 multimaximizer)
... (one or two minutes later) ... ok found it ...
lots of free vst plugins available so i will definitely not
PAY 40 bucks for a plugin ;-)
my used plugins (in order how audio runs) ...
reafir (surpress some noise ... especially some "rumble" on
the low end)
reaeq (shape the input signal a bit)
reagate (only send audio when you are speaking but low
background noise is surpressed)
reaxcomp (compressor with multiple bands) ... seems to be
similar to the l3
protoverb (just slight delay to give the audio a more filled
sound)
and finally frontier (self adaptive limiter) ... that
combination wortks superb BUT:
such a high number of plugins needs a lot of tweaking till
your audio sounds like from the bbc newsreader (grin) so do
not ask me how i have set it all ... it would take a much
longer mail to describe what i have done here
so finetune your audio (or making a complete mess from it)
can be done easy hi hi
the l3 multimaximizer seems to be similar to reaxcomp so i do
not need to try out the L3 i guess
greetz sigi dg9bfc
Am 31.08.2021 um 13:43 schrieb Conrad,
PA5Y:
Sigi as you are using VST plugins have
you tried the Waves L3 Multimaximizer? It is hard to make it
sound bad and it certainly increases talk power with minimal
distortion, it never overshoots.
73
Conrad
the nice thing when using all in software (sdr) you can do
things with compression, limiter, etc etc ... that you can
NOT do with analogue circuits
your "rf clipper" may produce a more boosted signal ... but
i guess when all is made in software you can be a bit better
the sdr has the audio in a chain (buffer) and you can have
a look forward agc ... limit every word to the maximum
without overshooting or harmonics produced
and with a monitor receiver on the output of your amp you
can add predistortion ... result is an even cleaner (but
more punchy) signal
no way to do that with conventional circuits
you could try to combine your rf clipper with the processor
in sdrc (but do not crank it up tooo much) ... i do a simlar
thingy with my mixing software (voicemeeter potatoe) with
some vst plugins added in the chain ... first the plugins do
a tiny bit of compression and limits the peaks not to
overdrive the sdrc input ... and a bit of processing of the
sdrc is added afterwards ... the result is a well boosted
signal with no splatter (pay also attention not to produce
intermod distortion on the final amp or driving stages!!)
simon you have the equalizer ... what if each filter in the
eq would have its own processor and limiter?? is that what
you mean with half octave filters (and boosting those) ?!?!?
dg9bfc sigi
Am 31.08.2021 um 12:01 schrieb MikeC:
Interesting article. I did experiment
further with the processor settings and compared the
result with an analogue speech processor that was part of
a ssb rig I designed about 20 years ago. This processor
uses variable rf log clipping of an ssb signal (at
10.7MHz) followed by another ssb filter. The audio is
pre-processed by a syllabic compressor (a vogad). The
clipping is done with two stages of log amplifiers. In the
original design the resulting ssb was converted to the
required frequency but I had already added a demodulator
back down to audio for something else so here the output
is audio. As the article points outs, non dc baseband
clipping avoids the serious harmonic distortion that
baseband clipping induces at all but very light clipping
because the harmonics multiply outside the passband. There
are still intermodulation products that eventually limit
performance.
Of course this processor is very
component hungry by today’s standards and I happened to
have it on the shelf. As the article also points out, the
the same result can be achieved in dsp by other means that
could not be realised in practice by analogue circuits.
I feed the audio output at line level
into the pc and have the SDRC processor turned off. The
audio processor bandwidth is from about 200Hz to 2.4kHz,
determined almost entirely by the analogue ssb filters.
I think that overshoot is a slightly
different, but related problem. Interestingly though I had
a look at this processor’s performance by putting test
sine and square-waves frequencies into it and looking at
the audio output on a scope. Probably by more luck than
judgement, the overshoot is a few percent unless the
processing is set ridiculously high. I suspect that a
fortuitous combination of filters and time constants, plus
the relatively narrow bandwidth requirement helps.
I have not checked for overshoot on the
rf output from the Limesdr that eventually generates the
signal for transmission.
My metering indicates that the average
to peak power ratio has increased significantly and,
although this is pretty unscientific, the “hf pileup
breaking index” is markedly improved.
Nothing can be all things to all men
(or women). Everything is a compromise of requirements,
performance, complexity, time and cost.
Mike
Sent from Mail for Windows
Probably.
I use a single band RMS compressor
followed by a very-well designed limiter which avoids
overshoot. I’ve tried multi-band compressors but not
achieved much benefit. When I get into my main winter
project I may look at this again, having a compressor
instance for each octave or half-octave.
For winter I want to finally play with
my Hermes-Lite 2, with my main amplifier I’ll have ~350
watts so will want to boost the audio.
Simon Brown, G4ELI
https://www.sdr-radio.com
Controlled-Envelope good?
http://www.arrl.org/files/file/QEX_Next_Issue/2014/Nov-Dec_2014/Hershberger_QEX_11_14.pdf
- 73 -
--
--
- + - + -
|
|
|
Gosh and I was so looking
forward to hearing you sing fiorentini schnell (I hope I got
the spelling right).
Gedas, W8BYA EN70JT
Gallery at http://w8bya.com (under repair)
Light travels faster than sound....
This is why some people appear bright until you hear them speak.
On 8/31/2021 3:24 PM, Siegfried
Jackstien wrote:
toggle quoted messageShow quoted text
hmm i have a good microphone ... wireless speakermike with ptt
... or in other words a mic with a 600 feet long cable hi hi
barefoot it sound "ok" (with some equalizer tweaking and a bit
of compression added) ... but with the vst plugins it is a bit
more punchy (without sounding bad)
mainly i installed the mixer cause i wanted to have a few
different audio ways for rx in parallel (drive speaker,
headphones and wireless mike plus maybe the bluetooth speakers
or the speakers fron the tv set that i use as big monitor for
sdrc)
the mixer on the tx part is also nice (easy to change levels
also for digimodes ...or whatever) ... the vst plugins just
added as "nice to have but not really needed"
no i will not take singing lessons (grin)
greetz sigi dg9bfc
Am 31.08.2021 um 17:45 schrieb Simon
Brown:
Or,
Just buy a decent microphone?
i do not know it ... (the l3 multimaximizer)
... (one or two minutes later) ... ok found it ...
lots of free vst plugins available so i will definitely not
PAY 40 bucks for a plugin ;-)
my used plugins (in order how audio runs) ...
reafir (surpress some noise ... especially some "rumble" on
the low end)
reaeq (shape the input signal a bit)
reagate (only send audio when you are speaking but low
background noise is surpressed)
reaxcomp (compressor with multiple bands) ... seems to be
similar to the l3
protoverb (just slight delay to give the audio a more
filled sound)
and finally frontier (self adaptive limiter) ... that
combination wortks superb BUT:
such a high number of plugins needs a lot of tweaking till
your audio sounds like from the bbc newsreader (grin) so do
not ask me how i have set it all ... it would take a much
longer mail to describe what i have done here
so finetune your audio (or making a complete mess from it)
can be done easy hi hi
the l3 multimaximizer seems to be similar to reaxcomp so i
do not need to try out the L3 i guess
greetz sigi dg9bfc
Am 31.08.2021 um 13:43 schrieb Conrad,
PA5Y:
Sigi as you are using VST plugins have
you tried the Waves L3 Multimaximizer? It is hard to make
it sound bad and it certainly increases talk power with
minimal distortion, it never overshoots.
73
Conrad
the nice thing when using all in software (sdr) you can
do things with compression, limiter, etc etc ... that you
can NOT do with analogue circuits
your "rf clipper" may produce a more boosted signal ...
but i guess when all is made in software you can be a bit
better
the sdr has the audio in a chain (buffer) and you can
have a look forward agc ... limit every word to the
maximum without overshooting or harmonics produced
and with a monitor receiver on the output of your amp you
can add predistortion ... result is an even cleaner (but
more punchy) signal
no way to do that with conventional circuits
you could try to combine your rf clipper with the
processor in sdrc (but do not crank it up tooo much) ... i
do a simlar thingy with my mixing software (voicemeeter
potatoe) with some vst plugins added in the chain ...
first the plugins do a tiny bit of compression and limits
the peaks not to overdrive the sdrc input ... and a bit of
processing of the sdrc is added afterwards ... the result
is a well boosted signal with no splatter (pay also
attention not to produce intermod distortion on the final
amp or driving stages!!)
simon you have the equalizer ... what if each filter in
the eq would have its own processor and limiter?? is that
what you mean with half octave filters (and boosting
those) ?!?!?
dg9bfc sigi
Am 31.08.2021 um 12:01 schrieb MikeC:
Interesting article. I did experiment
further with the processor settings and compared the
result with an analogue speech processor that was part
of a ssb rig I designed about 20 years ago. This
processor uses variable rf log clipping of an ssb signal
(at 10.7MHz) followed by another ssb filter. The audio
is pre-processed by a syllabic compressor (a vogad). The
clipping is done with two stages of log amplifiers. In
the original design the resulting ssb was converted to
the required frequency but I had already added a
demodulator back down to audio for something else so
here the output is audio. As the article points outs,
non dc baseband clipping avoids the serious harmonic
distortion that baseband clipping induces at all but
very light clipping because the harmonics multiply
outside the passband. There are still intermodulation
products that eventually limit performance.
Of course this processor is very
component hungry by today’s standards and I happened to
have it on the shelf. As the article also points out,
the the same result can be achieved in dsp by other
means that could not be realised in practice by analogue
circuits.
I feed the audio output at line level
into the pc and have the SDRC processor turned off. The
audio processor bandwidth is from about 200Hz to 2.4kHz,
determined almost entirely by the analogue ssb filters.
I think that overshoot is a slightly
different, but related problem. Interestingly though I
had a look at this processor’s performance by putting
test sine and square-waves frequencies into it and
looking at the audio output on a scope. Probably by more
luck than judgement, the overshoot is a few percent
unless the processing is set ridiculously high. I
suspect that a fortuitous combination of filters and
time constants, plus the relatively narrow bandwidth
requirement helps.
I have not checked for overshoot on
the rf output from the Limesdr that eventually generates
the signal for transmission.
My metering indicates that the
average to peak power ratio has increased significantly
and, although this is pretty unscientific, the “hf
pileup breaking index” is markedly improved.
Nothing can be all things to all men
(or women). Everything is a compromise of requirements,
performance, complexity, time and cost.
Mike
Sent from Mail for Windows
Probably.
I use a single band RMS compressor
followed by a very-well designed limiter which avoids
overshoot. I’ve tried multi-band compressors but not
achieved much benefit. When I get into my main winter
project I may look at this again, having a compressor
instance for each octave or half-octave.
For winter I want to finally play
with my Hermes-Lite 2, with my main amplifier I’ll have
~350 watts so will want to boost the audio.
Simon Brown, G4ELI
https://www.sdr-radio.com
Controlled-Envelope good?
http://www.arrl.org/files/file/QEX_Next_Issue/2014/Nov-Dec_2014/Hershberger_QEX_11_14.pdf
- 73 -
--
--
- + - + -
|
|
|
do re mi fa so la si doooooooo
Am 01.09.2021 um 00:09 schrieb Gedas:
toggle quoted messageShow quoted text
Gosh and I
was so looking forward to hearing you sing fiorentini
schnell (I hope I got the spelling right).
Gedas, W8BYA EN70JT
Gallery at http://w8bya.com (under repair)
Light travels faster than sound....
This is why some people appear bright until you hear them speak.
On 8/31/2021 3:24 PM, Siegfried
Jackstien wrote:
hmm i have a good microphone ... wireless speakermike with
ptt ... or in other words a mic with a 600 feet long cable hi
hi
barefoot it sound "ok" (with some equalizer tweaking and a
bit of compression added) ... but with the vst plugins it is a
bit more punchy (without sounding bad)
mainly i installed the mixer cause i wanted to have a few
different audio ways for rx in parallel (drive speaker,
headphones and wireless mike plus maybe the bluetooth speakers
or the speakers fron the tv set that i use as big monitor for
sdrc)
the mixer on the tx part is also nice (easy to change levels
also for digimodes ...or whatever) ... the vst plugins just
added as "nice to have but not really needed"
no i will not take singing lessons (grin)
greetz sigi dg9bfc
Am 31.08.2021 um 17:45 schrieb
Simon Brown:
Or,
Just buy a
decent microphone?
i do not know it ... (the l3 multimaximizer)
... (one or two minutes later) ... ok found it ...
lots of free vst plugins available so i will definitely
not PAY 40 bucks for a plugin ;-)
my used plugins (in order how audio runs) ...
reafir (surpress some noise ... especially some "rumble"
on the low end)
reaeq (shape the input signal a bit)
reagate (only send audio when you are speaking but low
background noise is surpressed)
reaxcomp (compressor with multiple bands) ... seems to be
similar to the l3
protoverb (just slight delay to give the audio a more
filled sound)
and finally frontier (self adaptive limiter) ... that
combination wortks superb BUT:
such a high number of plugins needs a lot of tweaking
till your audio sounds like from the bbc newsreader (grin)
so do not ask me how i have set it all ... it would take a
much longer mail to describe what i have done here
so finetune your audio (or making a complete mess from
it) can be done easy hi hi
the l3 multimaximizer seems to be similar to reaxcomp so
i do not need to try out the L3 i guess
greetz sigi dg9bfc
Am 31.08.2021 um 13:43 schrieb
Conrad, PA5Y:
Sigi as you are using VST plugins
have you tried the Waves L3 Multimaximizer? It is hard
to make it sound bad and it certainly increases talk
power with minimal distortion, it never overshoots.
73
Conrad
the nice thing when using all in software (sdr) you can
do things with compression, limiter, etc etc ... that
you can NOT do with analogue circuits
your "rf clipper" may produce a more boosted signal ...
but i guess when all is made in software you can be a
bit better
the sdr has the audio in a chain (buffer) and you can
have a look forward agc ... limit every word to the
maximum without overshooting or harmonics produced
and with a monitor receiver on the output of your amp
you can add predistortion ... result is an even cleaner
(but more punchy) signal
no way to do that with conventional circuits
you could try to combine your rf clipper with the
processor in sdrc (but do not crank it up tooo much) ...
i do a simlar thingy with my mixing software
(voicemeeter potatoe) with some vst plugins added in the
chain ... first the plugins do a tiny bit of compression
and limits the peaks not to overdrive the sdrc input ...
and a bit of processing of the sdrc is added afterwards
... the result is a well boosted signal with no splatter
(pay also attention not to produce intermod distortion
on the final amp or driving stages!!)
simon you have the equalizer ... what if each filter in
the eq would have its own processor and limiter?? is
that what you mean with half octave filters (and
boosting those) ?!?!?
dg9bfc sigi
Am 31.08.2021 um 12:01 schrieb
MikeC:
Interesting article. I did
experiment further with the processor settings and
compared the result with an analogue speech processor
that was part of a ssb rig I designed about 20 years
ago. This processor uses variable rf log clipping of
an ssb signal (at 10.7MHz) followed by another ssb
filter. The audio is pre-processed by a syllabic
compressor (a vogad). The clipping is done with two
stages of log amplifiers. In the original design the
resulting ssb was converted to the required frequency
but I had already added a demodulator back down to
audio for something else so here the output is audio.
As the article points outs, non dc baseband clipping
avoids the serious harmonic distortion that baseband
clipping induces at all but very light clipping
because the harmonics multiply outside the passband.
There are still intermodulation products that
eventually limit performance.
Of course this processor is very
component hungry by today’s standards and I happened
to have it on the shelf. As the article also points
out, the the same result can be achieved in dsp by
other means that could not be realised in practice by
analogue circuits.
I feed the audio output at line
level into the pc and have the SDRC processor turned
off. The audio processor bandwidth is from about 200Hz
to 2.4kHz, determined almost entirely by the analogue
ssb filters.
I think that overshoot is a
slightly different, but related problem. Interestingly
though I had a look at this processor’s performance by
putting test sine and square-waves frequencies into it
and looking at the audio output on a scope. Probably
by more luck than judgement, the overshoot is a few
percent unless the processing is set ridiculously
high. I suspect that a fortuitous combination of
filters and time constants, plus the relatively narrow
bandwidth requirement helps.
I have not checked for overshoot on
the rf output from the Limesdr that eventually
generates the signal for transmission.
My metering indicates that the
average to peak power ratio has increased
significantly and, although this is pretty
unscientific, the “hf pileup breaking index” is
markedly improved.
Nothing can be all things to all
men (or women). Everything is a compromise of
requirements, performance, complexity, time and cost.
Mike
Sent from Mail for Windows
Probably.
I use a single band RMS compressor
followed by a very-well designed limiter which avoids
overshoot. I’ve tried multi-band compressors but not
achieved much benefit. When I get into my main winter
project I may look at this again, having a compressor
instance for each octave or half-octave.
For winter I want to finally play
with my Hermes-Lite 2, with my main amplifier I’ll
have ~350 watts so will want to boost the audio.
Simon Brown, G4ELI
https://www.sdr-radio.com
Controlled-Envelope good?
http://www.arrl.org/files/file/QEX_Next_Issue/2014/Nov-Dec_2014/Hershberger_QEX_11_14.pdf
- 73 -
--
--
- + - + -
|
|
|
I agree buy a good microphone, although it need not be that good for amateur SSB and NBFM radio use. However a good microphone will not help you have punchy audio.
Having a good microphone AND using high quality audio processing is not mutually exclusive. If you want really well behaved punchy audio from DXer to BBC presenter and are already using VST plugins then the L3 multimaximizer is a very good
easy to use one stop solution. Waves plug ins have been in regular use in broadcasting for a long time.
Sigi you need to compare the L3 to your free plug ins to appreciate the difference. I would NEVER use any reverb processor, you really don’t need it, in fact it almost certainly will reduce intelligibility.
Personally I am perfectly happy with the processing on my TS-890S or K3S. The 890S sounds very natural but louder than unprocessed audio.
Conrad
toggle quoted messageShow quoted text
From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io>
On Behalf Of Simon Brown via groups.io
Sent: 31 August 2021 17:45
To: main@SDR-Radio.groups.io
Subject: Re: [SDR-Radio] Speech Processor
Or,
Just buy a decent microphone?
i do not know it ... (the l3 multimaximizer)
... (one or two minutes later) ... ok found it ...
lots of free vst plugins available so i will definitely not PAY 40 bucks for a plugin ;-)
my used plugins (in order how audio runs) ...
reafir (surpress some noise ... especially some "rumble" on the low end)
reaeq (shape the input signal a bit)
reagate (only send audio when you are speaking but low background noise is surpressed)
reaxcomp (compressor with multiple bands) ... seems to be similar to the l3
protoverb (just slight delay to give the audio a more filled sound)
and finally frontier (self adaptive limiter) ... that combination wortks superb BUT:
such a high number of plugins needs a lot of tweaking till your audio sounds like from the bbc newsreader (grin) so do not ask me how i have set it all ... it would take a much longer mail to describe what i have done here
so finetune your audio (or making a complete mess from it) can be done easy hi hi
the l3 multimaximizer seems to be similar to reaxcomp so i do not need to try out the L3 i guess
greetz sigi dg9bfc
Am 31.08.2021 um 13:43 schrieb Conrad, PA5Y:
Sigi as you are using VST plugins have you tried the Waves L3 Multimaximizer? It is hard to make it sound bad and it certainly increases talk power with minimal distortion, it never overshoots.
73
Conrad
the nice thing when using all in software (sdr) you can do things with compression, limiter, etc etc ... that you can NOT do with analogue circuits
your "rf clipper" may produce a more boosted signal ... but i guess when all is made in software you can be a bit better
the sdr has the audio in a chain (buffer) and you can have a look forward agc ... limit every word to the maximum without overshooting or harmonics produced
and with a monitor receiver on the output of your amp you can add predistortion ... result is an even cleaner (but more punchy) signal
no way to do that with conventional circuits
you could try to combine your rf clipper with the processor in sdrc (but do not crank it up tooo much) ... i do a simlar thingy with my mixing software (voicemeeter potatoe) with some vst plugins added in the chain ... first the plugins
do a tiny bit of compression and limits the peaks not to overdrive the sdrc input ... and a bit of processing of the sdrc is added afterwards ... the result is a well boosted signal with no splatter (pay also attention not to produce intermod distortion on
the final amp or driving stages!!)
simon you have the equalizer ... what if each filter in the eq would have its own processor and limiter?? is that what you mean with half octave filters (and boosting those) ?!?!?
dg9bfc sigi
Am 31.08.2021 um 12:01 schrieb MikeC:
Interesting article. I did experiment further with the processor settings and compared the result with an analogue speech processor that was part of a ssb rig I designed about 20 years ago. This processor uses variable
rf log clipping of an ssb signal (at 10.7MHz) followed by another ssb filter. The audio is pre-processed by a syllabic compressor (a vogad). The clipping is done with two stages of log amplifiers. In the original design the resulting ssb was converted to the
required frequency but I had already added a demodulator back down to audio for something else so here the output is audio. As the article points outs, non dc baseband clipping avoids the serious harmonic distortion that baseband clipping induces at all but
very light clipping because the harmonics multiply outside the passband. There are still intermodulation products that eventually limit performance.
Of course this processor is very component hungry by today’s standards and I happened to have it on the shelf. As the article also points out, the the same result can be achieved in dsp by other means that could not be
realised in practice by analogue circuits.
I feed the audio output at line level into the pc and have the SDRC processor turned off. The audio processor bandwidth is from about 200Hz to 2.4kHz, determined almost entirely by the analogue ssb filters.
I think that overshoot is a slightly different, but related problem. Interestingly though I had a look at this processor’s performance by putting test sine and square-waves frequencies into it and looking at the audio
output on a scope. Probably by more luck than judgement, the overshoot is a few percent unless the processing is set ridiculously high. I suspect that a fortuitous combination of filters and time constants, plus the relatively narrow bandwidth requirement
helps.
I have not checked for overshoot on the rf output from the Limesdr that eventually generates the signal for transmission.
My metering indicates that the average to peak power ratio has increased significantly and, although this is pretty unscientific, the “hf pileup breaking index” is markedly improved.
Nothing can be all things to all men (or women). Everything is a compromise of requirements, performance, complexity, time and cost.
Mike
Sent from
Mail for Windows
Probably.
I use a single band RMS compressor followed by a very-well designed limiter which avoids overshoot. I’ve tried multi-band compressors but not achieved much benefit. When I get into my main winter project I may look at
this again, having a compressor instance for each octave or half-octave.
For winter I want to finally play with my Hermes-Lite 2, with my main amplifier I’ll have ~350 watts so will want to boost the audio.
Simon Brown, G4ELI
https://www.sdr-radio.com
Controlled-Envelope good?
http://www.arrl.org/files/file/QEX_Next_Issue/2014/Nov-Dec_2014/Hershberger_QEX_11_14.pdf
- 73 -
--
--
|
|
|
Reverb? That's the "Good buddy" sound. A relatively
flat spectrum with a broad notch at about 1.2 kHz, depending on
the voice, concentrates your power where it does good. For DX and
contests reduce the bass, too. For ad hoc on the air chats over a
virtual cafe table raise the bass no more than 6dB. Going for the
voice of God sound with or without the reverb and other "stuff"
tell me what I've run across is not people I want to talk with.
{^_-}
toggle quoted messageShow quoted text
On 20210901 08:19:41, Conrad, PA5Y
wrote:
I agree buy a good microphone, although it
need not be that good for amateur SSB and NBFM radio use.
However a good microphone will not help you have punchy audio.
Having a good microphone AND using high
quality audio processing is not mutually exclusive. If you
want really well behaved punchy audio from DXer to BBC
presenter and are already using VST plugins then the L3
multimaximizer is a very good easy to use one stop solution.
Waves plug ins have been in regular use in broadcasting for a
long time.
Sigi you need to compare the L3 to your
free plug ins to appreciate the difference. I would NEVER use
any reverb processor, you really don’t need it, in fact it
almost certainly will reduce intelligibility.
Personally I am perfectly happy with the
processing on my TS-890S or K3S. The 890S sounds very natural
but louder than unprocessed audio.
Conrad
Or,
Just buy a decent microphone?
i do not know it ... (the l3
multimaximizer)
... (one or two minutes later) ... ok
found it ...
lots of free vst plugins available so i
will definitely not PAY 40 bucks for a plugin ;-)
my used plugins (in order how audio runs)
...
reafir (surpress some noise ... especially
some "rumble" on the low end)
reaeq (shape the input signal a bit)
reagate (only send audio when you are
speaking but low background noise is surpressed)
reaxcomp (compressor with multiple bands)
... seems to be similar to the l3
protoverb (just slight delay to give the
audio a more filled sound)
and finally frontier (self adaptive
limiter) ... that combination wortks superb BUT:
such a high number of plugins needs a lot
of tweaking till your audio sounds like from the bbc
newsreader (grin) so do not ask me how i have set it all ...
it would take a much longer mail to describe what i have
done here
so finetune your audio (or making a
complete mess from it) can be done easy hi hi
the l3 multimaximizer seems to be similar
to reaxcomp so i do not need to try out the L3 i guess
greetz sigi dg9bfc
Am 31.08.2021 um 13:43
schrieb Conrad, PA5Y:
Sigi as you are using
VST plugins have you tried the Waves L3 Multimaximizer? It
is hard to make it sound bad and it certainly increases
talk power with minimal distortion, it never overshoots.
73
Conrad
the nice thing when using all in
software (sdr) you can do things with compression,
limiter, etc etc ... that you can NOT do with analogue
circuits
your "rf clipper" may produce a more
boosted signal ... but i guess when all is made in
software you can be a bit better
the sdr has the audio in a chain
(buffer) and you can have a look forward agc ... limit
every word to the maximum without overshooting or
harmonics produced
and with a monitor receiver on the
output of your amp you can add predistortion ... result is
an even cleaner (but more punchy) signal
no way to do that with conventional
circuits
you could try to combine your rf clipper
with the processor in sdrc (but do not crank it up tooo
much) ... i do a simlar thingy with my mixing software
(voicemeeter potatoe) with some vst plugins added in the
chain ... first the plugins do a tiny bit of compression
and limits the peaks not to overdrive the sdrc input ...
and a bit of processing of the sdrc is added afterwards
... the result is a well boosted signal with no splatter
(pay also attention not to produce intermod distortion on
the final amp or driving stages!!)
simon you have the equalizer ... what if
each filter in the eq would have its own processor and
limiter?? is that what you mean with half octave filters
(and boosting those) ?!?!?
dg9bfc sigi
Am 31.08.2021 um
12:01 schrieb MikeC:
Interesting article.
I did experiment further with the processor settings and
compared the result with an analogue speech processor
that was part of a ssb rig I designed about 20 years
ago. This processor uses variable rf log clipping of an
ssb signal (at 10.7MHz) followed by another ssb filter.
The audio is pre-processed by a syllabic compressor (a
vogad). The clipping is done with two stages of log
amplifiers. In the original design the resulting ssb was
converted to the required frequency but I had already
added a demodulator back down to audio for something
else so here the output is audio. As the article points
outs, non dc baseband clipping avoids the serious
harmonic distortion that baseband clipping induces at
all but very light clipping because the harmonics
multiply outside the passband. There are still
intermodulation products that eventually limit
performance.
Of course this
processor is very component hungry by today’s standards
and I happened to have it on the shelf. As the article
also points out, the the same result can be achieved in
dsp by other means that could not be realised in
practice by analogue circuits.
I feed the audio
output at line level into the pc and have the SDRC
processor turned off. The audio processor bandwidth is
from about 200Hz to 2.4kHz, determined almost entirely
by the analogue ssb filters.
I think that
overshoot is a slightly different, but related problem.
Interestingly though I had a look at this processor’s
performance by putting test sine and square-waves
frequencies into it and looking at the audio output on a
scope. Probably by more luck than judgement, the
overshoot is a few percent unless the processing is set
ridiculously high. I suspect that a fortuitous
combination of filters and time constants, plus the
relatively narrow bandwidth requirement helps.
I have not checked
for overshoot on the rf output from the Limesdr that
eventually generates the signal for transmission.
My metering
indicates that the average to peak power ratio has
increased significantly and, although this is pretty
unscientific, the “hf pileup breaking index” is markedly
improved.
Nothing can be all
things to all men (or women). Everything is a compromise
of requirements, performance, complexity, time and cost.
Mike
Sent from
Mail for Windows
Probably.
I use a single band
RMS compressor followed by a very-well designed limiter
which avoids overshoot. I’ve tried multi-band
compressors but not achieved much benefit. When I get
into my main winter project I may look at this again,
having a compressor instance for each octave or
half-octave.
For winter I want to
finally play with my Hermes-Lite 2, with my main
amplifier I’ll have ~350 watts so will want to boost the
audio.
Simon Brown, G4ELI
https://www.sdr-radio.com
Controlled-Envelope
good?
http://www.arrl.org/files/file/QEX_Next_Issue/2014/Nov-Dec_2014/Hershberger_QEX_11_14.pdf
- 73 -
--
--
|
|
|
