Closed tonyc closed 10 months ago
This would be a really excellent feature addition Tony, I hope you are able to consider it soon.
I have the basics of this working in spike/audio-tx, but there is a lot of cleanup to do. I suspect I will have this shipped by the end of this year, though.
Is it worth having a go with this now @tonyc or perhaps wait until you have done more cleaning up?
I'd say wait a little bit until I get it cleaned up. Once it is usable, I will just merge it straight into main
.
There are a couple extra buttons in the UI that I was using for debugging, sort of like a manual PTT, because I wasn't sure how to get it working with the radio's SEND button, but I've got that figured out now.
Surprisingly, once I started to dig into it, it wasn't all that hard of a problem.. I probably could have had this working a couple years ago 😆 The main issue I ran into was realizing that the audio (both in and out) aren't 8-bit values, but 16-bit unsigned values. So I ended up needing fewer samples per UDP packet, but then I needed to figure out how to split e.g. a 16-bit unsigned value into its two separate bytes, to stuff into the UDP packet.
The other problem I ran into was with loudness. The JS audio library I'm using gives me sample values in the range of -32768 ... 32767
, and the "loud" values of the audio samples were way too loud, so I had to scale them all by .002 to not blow out the audio.
Once I got the byte formatting and the audio volume figured out, it all pretty much just worked - and I had a SSB contact using the app earlier today!
OK, I will watch with interest and see how things progress.
@Tyrbiter I just had my second-ever SSB QSO with open890 on this branch, feel free to pull it down and give it a shot.
You will need to hold the DATA button and set up the rear SEND/PTT like this:
This config will let you use the SEND button to transmit.
Make sure DATA VOX is off, otherwise it will operate with VOX, which you probably don't want.
Click the "VOIP MIC: OFF" button at the top, and your browser will prompt you for permission. Make sure to choose the correct audio device, otherwise if you accidentally choose the TS-890's USB audio device, you'll create some great feedback:
You may need to reduce the KNS audio level, and also your OS-level mic level.
The KNS config is at: MENU -> KNS -> CONFIG -> Audio Input Level (VOIP)
I've got mine set to 25, and it seems to give good results.
Actually, I've just merged this into main
and deleted the branch. I have a couple of other QOL cleanup things to do, but this is mostly finished.
Just saw this, I will have a play and see what I find. I think this is where a macro or two will be needed to set up the audio paths, I currently use audio over USB using freedv in analog voice mode.
Thanks for all the work!
Is there a way to capture PCM samples in the browser?
Required for full remote TX operation