Transmit Audio Monitoring Suggestions

W8JJ
Posts: 19
Joined: Wed Dec 16, 2020 3:18 am

Transmit Audio Monitoring Suggestions

Postby W8JJ » Wed Mar 31, 2021 4:29 pm

I'm interested in monitoring my audio during transmit using headphones. However, the MON feature seems unusable to me as currently configured due to latency issues. Is there a way to fix this via software settings? Please note that I'm NOT referring to using the onboard monitoring circuit for audio set up and equalization. Rather, I'm interested in monitoring my voice to better control proximity effect, tone, pitch, volume, and my rate of speech. I often create online content and live stream and I've come to realize the value of real time audio monitoring.

Please share your thoughts and techniques for effective everyday audio monitoring AFTER levels and EQ are set up using an external receiver.

7000DLE MKII, Thetis 2.8.1, i7 9700, 16GB RAM, Win10 Professional

73 Tim W8JJ
User avatar
w-u-2-o
Posts: 5525
Joined: Fri Mar 10, 2017 1:47 pm

Re: Transmit Audio Monitoring Suggestions

Postby w-u-2-o » Wed Mar 31, 2021 5:20 pm

Unless you are one of those humans who is immune to listening to an echo of themselves, it is effectively impossible to do what you want because of the latencies (delays) inherent in the openHPSDR software architecture.

If you eliminate all audio processing in Thetis/PowerSDR: limiter, pre and post EQ, CFC, expander/gate, etc., and pay very, very strict attention to buffer settings, use VAC with ASIO, essentially every trick in the book, you can get the mic to antenna delay down to around 30ms, which is already in the noticeable range. Add back in all the audio processing delays and you start to get up around 100ms, which is well into the plain old echo range.

Hence monitoring with an external receiver will suffer the same problems as MON, although MON has another fatal flaw: it is tapped off after, not before, PureSignal predistortion is added, which makes it sound awful even if it did not have latency issues.

Real time audio monitoring of processed audio has been a problem since the advent of digital audio processing. If you are using a DAW with a lot of plug-ins you are again faced with the same issue. For broadcasters using something like an Orban or an OMNIA unit, these create many 100's of milliseconds of delay, i.e. same problems but even worse.

Solutions to this problem include:

a) Using a recorded loop of your unprocessed speech to make adjustments to processing, this way you are only listening to the finished product.
b) Real time analog monitoring prior to processing (or a low latency unprocessed digital send) as an adjunct to (a).
c) Throwing in the towel on digital processing and using a rack-load of old analog equipment.

73,

Scott
User avatar
KX4M
Posts: 8
Joined: Tue Dec 29, 2020 7:46 pm

Re: Transmit Audio Monitoring Suggestions

Postby KX4M » Wed Mar 31, 2021 6:23 pm

One possible solution that I use in other applications is the Direct Monitor feature on Focusrite’s Scarlet series of audio interfaces. This will allow you to monitor yourself in low latency via the headphone out on the device.

With this method you will hear things like proximity effect and plosives, but not compression, expanders, and EQ applied later in your digital audio chain.
User avatar
w-u-2-o
Posts: 5525
Joined: Fri Mar 10, 2017 1:47 pm

Re: Transmit Audio Monitoring Suggestions

Postby w-u-2-o » Wed Mar 31, 2021 7:04 pm

KX4M wrote:One possible solution that I use in other applications is the Direct Monitor feature on Focusrite’s Scarlet series of audio interfaces. This will allow you to monitor yourself in low latency via the headphone out on the device.

That's my (b) solution, above.
User avatar
kc2rgw
Posts: 165
Joined: Mon Jun 22, 2020 5:44 pm

Re: Transmit Audio Monitoring Suggestions

Postby kc2rgw » Thu Apr 01, 2021 4:09 am

I use two ways to do it.

A multi port switch with a dummy load on the COM. Anan on the active port and a receiver on the 2nd port. With the typical Alpha Delta switch, there is plenty of leakage on the switch for the 2nd receiver.

I use an RSPdx from SDR Play and SDRConsole with it. It has simple recording built in and flexible filtering to open up for 5kHz of audio and good SAM mode for setting up for AM.

The other way, since I got rid of the Alpha Delta switches and went to BNC patch bay. The patch bay has no leakage so now I use an RF sampler/coupler on a coax to a dummy load. I transmit into the dummy load and feed the output of the RF sampler to the 2nd receiver.

With 15-20W my RF sampler feeds about a 30-50 over S9 signal level. Just be cautious doing this to find how hot the signal out of the coupler will be for a given TX power level.

It is really key to record and play back as listening to your own audio “live” will result in poor perception due to bone conductance in your head while talking. Typically you wind up with far too much bass if you monitor yourself live. You can get close live, but proof it with a record/play cycle.
JJ4SDR
Posts: 434
Joined: Fri Jul 30, 2021 10:09 pm
Location: TEXAS, USA

Re: Transmit Audio Monitoring Suggestions

Postby JJ4SDR » Thu Aug 05, 2021 9:43 pm

KX4M wrote:One possible solution that I use in other applications is the Direct Monitor feature on Focusrite’s Scarlet series of audio interfaces. This will allow you to monitor yourself in low latency via the headphone out on the device.

With this method you will hear things like proximity effect and plosives, but not compression, expanders, and EQ applied later in your digital audio chain.


Tapping into the headphones OUT signal on the Scarlet: Isn’t what you are hearing actually something much wider (perhaps approaching or even exceeding 20 kHz) in frequency band-width than what you actually sound like? That OUT circuit is really working as it should when you are in a recording studio right? Most of what you hear through that circuit will not be transmitted or used to modulate an SSB signal. Perhaps your “other application” is studio recordings or other entirely non-ham radio related application?

Best!
JJ
PC: 8 Core i7-10700 CPU @ 2.90GHz, NVMe SK Hynix 512 GB SSD, 32GB RAM
Windows 10 Home, Version 22H2
Thetis v2.10.4.3 x64
Protocol 2 v2.1.18
JJ4SDR
Posts: 434
Joined: Fri Jul 30, 2021 10:09 pm
Location: TEXAS, USA

Re: Transmit Audio Monitoring Suggestions

Postby JJ4SDR » Thu Aug 05, 2021 11:37 pm

JJ4SDR wrote:
KX4M wrote:One possible solution that I use in other applications is the Direct Monitor feature on Focusrite’s Scarlet series of audio interfaces. This will allow you to monitor yourself in low latency via the headphone out on the device.

With this method you will hear things like proximity effect and plosives, but not compression, expanders, and EQ applied later in your digital audio chain.


Tapping into the headphones OUT signal on the Scarlet: Isn’t what you are hearing actually something much wider (perhaps approaching or even exceeding 20 kHz) in frequency band-width than what you actually sound like? That OUT circuit is really working as it should when you are in a recording studio right? Most of what you hear through that circuit will not be transmitted or used to modulate an SSB signal. Perhaps your “other application” is studio recordings or other entirely non-ham radio related application?

Best!
JJ


Add by JJ:
Also, what you hear through Scarlet isn’t in RF domain but AF domain as well.
PC: 8 Core i7-10700 CPU @ 2.90GHz, NVMe SK Hynix 512 GB SSD, 32GB RAM
Windows 10 Home, Version 22H2
Thetis v2.10.4.3 x64
Protocol 2 v2.1.18
User avatar
w-u-2-o
Posts: 5525
Joined: Fri Mar 10, 2017 1:47 pm

Re: Transmit Audio Monitoring Suggestions

Postby w-u-2-o » Fri Aug 06, 2021 12:01 am

JJ,

As someone new to the community you might not realize that that issue was intuitively obvious in the discussion above.

Bottom line: you can NOT do real-time monitoring of transmitted audio in the openHPSDR architecture. There is simply too much latency/delay. All you can do is make a recording of some source material and play that into Thetis so you can only listen to the output as you make your adjustments.

You can use the monitor output of something like a Focusrite to provide unprocessed sidetone, that's about it.

73,

Scott

Return to “Microphones, Speakers & Audio Hardware”