Re: Happy New Year


Simon Brown
 

Agreed about the training. I hope to have the AI included by Easter, then I can dig further. Still a lot to do to get a nice UI harness working.

 

Simon Brown, G4ELI

https://www.sdr-radio.com

 

From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of Milo Medin (K6CZ)
Sent: 02 January 2022 16:31
To: main@sdr-radio.groups.io
Cc: SDR-Radio@groups.io
Subject: Re: [SDR-Radio] Happy New Year

 

Tensorflow is an excellent system for this sort of thing.  The author of that article used random words to train the model, but if he used more traditional ham conversations it would help a lot, at least for amatuer use including typical abbreviations and Q signs so they would be better represented and not read as errors by the model.  The other thing that would be useful would be to send a bunch of  known CW text from an automated keyer and pick up the audio from a remote SDR, so the model would include fading and other non-linear impairments that come from real HF paths that hams deal with all the time.  The cherry on top would be to have hams send a set of known text with keyer, bug or straight key and capture that over the HF signal path as well.  

 

If the ML model is trained well, I would not be at all surprised if it could decode CW much better than the best human.  I have been surprised as to how amazing the pattern recognition capabilities of well trained models are.  The key is all in the training.  

 

Good luck,

Milo (K6CZ)

 

 

 

On Sat, Jan 1, 2022 at 1:14 PM Simon Brown <simon@...> wrote:

Milo,

 

And excellent article – something I want to understand this year.

 

Simon Brown, G4ELI

https://www.sdr-radio.com

 

From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of Milo Medin (K6CZ)
Sent: 01 January 2022 20:22
To: main@sdr-radio.groups.io
Cc: SDR-Radio@groups.io
Subject: Re: [SDR-Radio] Happy New Year

 

There are a huge number of new tools for using AI today that leverage modern compute infrastructure.  Unfortunately, many of the public ML tools out there are optimized for visual image processing, as opposed to audio or RF SIGINT style uses.  But a number of efforts exist to convert audio or RF to an image, and then use the image tools to do the training and generate a model for decoding.  While the training is computationally expensive, once the model is built, it can be pretty efficient to run.  

 

Here is one that is aimed at the cw decoding problem specifically that outlines use of some common AI tools: http://ag1le.blogspot.com/2019/02/training-computer-to-listen-and-decode.html

 

There was another one that used ML,  GIMP and RTL-SDR's to do signal identification using images again, but I can't seem to find it now.  

 

Good luck,

milo (K6CZ)

 

On Sat, Jan 1, 2022 at 9:55 AM Simon Brown <simon@...> wrote:

Well,

 

This actually will be a step into AI in 2023 – 1988 / 1989 I was working in AI at SWIFT, sadly the brains of the operation was on the plane that crashed at Lockerbie. 30+ years later it’s time to try and remember the theory and see how it’s changed.

 

Simon Brown, G4ELI

https://www.sdr-radio.com

 

From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of Larry Dodd
Sent: 01 January 2022 17:33
To: main@sdr-radio.groups.io
Cc: SDR-Radio@groups.io
Subject: Re: [SDR-Radio] Happy New Year

 

Simon

You will do a great job. Solving the unsteady CW hand issue will be interesting. Even some machine produced CW will be a challenge to decode. Looking forward to your work on it in a few months from now. Computers are amazing but programmers even more amazing. Ha!

Larry K4LED 

 

On Jan 1, 2022, at 11:35 AM, Simon Brown <simon@...> wrote:



Hi,

 

I’ve often said that a computer can decode better than the human brain, now I’m trying to prove it. ‘I may be some time…’

 

Simon Brown, G4ELI

https://www.sdr-radio.com

 

From: main@SDR-Radio.groups.io <main@SDR-Radio.groups.io> On Behalf Of Larry Dodd
Sent: 01 January 2022 15:13
To: main@sdr-radio.groups.io
Cc: SDR-Radio@groups.io
Subject: Re: [SDR-Radio] Happy New Year

 

Simon. All,

Happy 2022 to all. May good SDR things happen this year. Good luck Simon on the CW decoder. It’s much needed. 

Larry K4LED 

 

On Jan 1, 2022, at 9:02 AM, Simon Brown <simon@...> wrote:



Hi All,

 

Over here it’s 2022 so a very happy new year to you all. I’ve updated the current V3.1 beta and am now spending January designing a CW Decoder harness which I hope will help me develop a good CW Decoder, so for January and possibly February you’ll not be hearing much from me 😊 .

 

Be good in 2022,

 

Simon Brown, G4ELI

https://www.sdr-radio.com

 


--

- + - + -

Please use https://forum.sdr-radio.com:4499/ when posting questions or problems.


--

- + - + -

Please use https://forum.sdr-radio.com:4499/ when posting questions or problems.


--

- + - + -

Please use https://forum.sdr-radio.com:4499/ when posting questions or problems.


--

- + - + -

Please use https://forum.sdr-radio.com:4499/ when posting questions or problems.


--
- + - + -
Please use https://forum.sdr-radio.com:4499/ when posting questions or problems.

Join main@SDR-Radio.groups.io to automatically receive all group messages.