new improved python qrq cw keyboard from grok3 AI

some improvements have been issued from GROK3 on this latest QRQ CW KEYBOARD

written in PYTHON code and the multiple cw timing adjustments were drawn from YPLOG's CW KEYBOARDs multiple cw timing settings ,   

the python code file can be downloaded if you want to test it...

uploaded/attach it to your favorite AI and ask it to convert the audio engine over to windows or mac...this is written for the LINUX OS

https://u.pcloud.link/publink/show?code=XZYLC45Zo6kjF6AJQ3jc7eVsuCBz2jK7bjnk

looks like this:

Load Previous Comments
  • Thomas MARTIN

    My three favorite shapes for listening to e-books are:

    (Ramp Duration = 7 ms, Compensation = 2.5 ms)

    1.) "Bartlett-Hann"

    2.) "SigmoidGaussian"

    3.) "Gudermannian"

    TNX Chuck for all the time and effort invested to make the Python Morse Code Keyboard!

    73

    Tom DF7TV

  • Thomas MARTIN

    That's how I have found my favorite shape:

    --

    Python Morse Code Keyboard, Favorites        DF7TV MAR 31, 2025

    Frequency: 700 Hz, Volume: 0,2

    All other parameters = default values

    (Text: N0HFF ASRT)

    Most pleasant sound (unordered):

    ================================

    + "Hann"

    + "Bartlett-Hann"

    + CosineSquared"

    + "Sigmoid(a=2)"

    + "Gudermannian"

    + "SigmoidTapered"

    + "SigmoidGaussian"

    + "PSWF"

    Readability → less = -1, okay = 0, better = +1

    ==============================================

    0     "Hann"

    +1   "Bartlett-Hann"

    0     "CosineSquared"

    +1   "Sigmoid(a=2)"

    +1   "Gudermannian"

    0     "SigmoidTapered"

    +1   "SigmoidGaussian"

    +1   "PSWF"

    Reduce to the three best ones

    =============================

    1.)  "Bartlett-Hann"

    2.)  "SigmoidGaussian"

    3.)  "Gudermannian"

    Optimize parameters for winner BARTLETT-HANN

    ============================================

    Ramp Duration: 7 ms

    Compensation: 2.5 ms

    --

    73

    Tom DF7TV

  • Chuck aa0hw

    Here is a comparison graph between Raised Cosine and PSWF function both set to 8ms rise/fall time...  look at the side lobes of noise each makes and notice that the PSWF has   30 DB LESS  side lobe noise close in to the main cw pitch than the RC does...