Multichannel Audio Recording

🙋 If you want a new feature or a change to the software then add your request here.

If it's not here then it's not going to happen.
pauldbnut
Posts: 115
Joined: Wed Aug 19, 2020 10:59 am
Location: Dinosaur Country (IoW)

Multichannel Audio Recording

#1

Unread post by pauldbnut »

At the moment SDR Console can "Record Active Receiver" to a standard WAV, WMA or MP3.

I would be very interested in a feature allowing me to make a synchronised recording of two RX simultaneously (say one RX to Left, the other to Right). This would be used for monitoring broadcast stations on parallel channels, and have the double benefit of reviewing a session for periods of clear overlapping reception plus submitting clips in a single file as evidence.

Why stop at two? In principle, a monitoring session could be expanded to a large number of audio streams and not limited to just parallel broadcast channels. The WAV format can theoretically handle up to 64K channels, and there is player/editor software available to review and edit individual channels or combinations. (I realise the monitored channels would need to be within the same sampled bandwidth.)

The only downside is that MP3 compression is not available for more than 2 channels. WMA has the same limitation, although WMA Pro can handle 8 channels. That would not be a problem for me.

Would it be possible to include periodic timestamp metadata, rather than rely on file title? I've not looked at this for several years and cannot remember how you managed it for I-Q recordings.
Paul White (Grumpy Old Git)

pauldbnut
Posts: 115
Joined: Wed Aug 19, 2020 10:59 am
Location: Dinosaur Country (IoW)

Re: Multichannel Audio Recording

#2

Unread post by pauldbnut »

Here is a sample recording of Taiwan Area Fisheries Broadcast via Bjarne's Kongsfjord KiwiSDR.
Streamed using John Seamons' KiwiRecorder tool on two simultaneous channels for 738 & 1433 kHz, broadcasting in parallel.
The two mono streams were stitched into a single stereo WAV with Audacity.
https://1drv.ms/u/s!Aowxw4jVL-1TdtFPiJSHgzcS9rc
I think it's a great way to verify station ID when parallel outlets are available.
Paul White (Grumpy Old Git)

User avatar
Simon G4ELI
Posts: 2134
Joined: Thu Aug 06, 2020 7:27 am
Location: Mawnan Smith
Contact:

Re: Multichannel Audio Recording

#3

Unread post by Simon G4ELI »

Paul,

Were I to make a multi-channel recording, how would you play it? IMO just make an IQ recording, then you have everything.

As for timestamps, this may be possible with WMA but isn't with MP3 / WAV as far as I know - it's up to the audio player to work this out.
Simon Brown :shock:
www.sdr-radio.com

Do not send me direct e-mail, thank-you!

pauldbnut
Posts: 115
Joined: Wed Aug 19, 2020 10:59 am
Location: Dinosaur Country (IoW)

Re: Multichannel Audio Recording

#4

Unread post by pauldbnut »

Hi, Simon, hope Xmas is good for you & hairy.
I admit the >2 channels was more "let's see where this goes" than a request that could easily be justified. But:
  • Reaper and any number of DAW/Editors could replay them.
  • A broadcaster (asked for QSL) would probably have no trouble either.
  • It's the perfect archive format for exactly those parallel channels received in a given session.
  • In almost every case, it's a tiny fraction the size of a "containing" IQ.
  • IQs are not suitable as QSL evidence, or for sharing on forums etc.
That said, now you've brought me down to earth, multi (>2) really is hard to justify.
Besides, even if you provided a player for them, L/R selection/comparison gets confusing.

Agreed that a simple RXA/RXB -> L/R would be:
  • So much easier to understand.
  • Playable by existing audio replay (yours and any other).
  • And... greatly assisted by a "front-of-house" L/R balance control.
I really appreciate your scrutiny and feedback.
May I please scale back the request to just RXA/RXB (or whatever) "stereo"?

As for implementation/GUI, I'm out of my depth.
It would have to be simple & fool-proof to switch between independent RX1-n and paired RXA/RXB modes (maybe there's the clue?).

OK.

Timestamps continue to haunt me:
  • They should be possible with BWF chunks.
  • They would be preserved independently of filename changes or editing (e.g. samples from a longer file).
  • (so might be useful for IQ as well?)
Maybe it's better I have a look at Kiwi/GPS for a better understanding of the possibilities.
So let's leave timestamps until another time, shall we?

Thanks for everything.
Paul White (Grumpy Old Git)

Post Reply