I have tried many things to keep my hands reasonably warm during CW activations.
When I first saw the crazy idea of using voice instead of a key, I couldn’t resist…
This is a very first test from home.
EDIT 2023/03/15: This is about keying a CW transmitter using voice, NOT about sending dits and dahs over SSB. Please listen carefully.
I guess the main difficulty will be to reduce the wind noise to an acceptable level so that voice can be properly detected above noise to trigger properly.
IMO: If you mean putting your rig in SSB mode, and telling dits or dahs via a mic, it is Morse code but not CW. Genuine CW is simply putting off/on a carrier to transmit the dits or dahs. The ‘mic’ mode you are describing is called MCW, very inefficient and it generates lots IMDs. It is not legal in USA for amateur radio.
As a person almost congenitally incapable of doing Morse, I’m a little disappointed to see that this new Arduino project, good though it undoubtedly is, does not recognize human speech - e.g. “see cue SOTA dee em one see em portable”, etc. - and produce a CW signal from that. I realize I’m expecting rather too much at the moment - I don’t know if there are any speech-recognition “chips” which might be incorporated in such a thing?
Well, this would be quite a drift from the original idea.
I thought the CWvox project was crazy enough to be fun to experiment with, and incidentally, it could help keep my hands warm for some activations.
I did not use the original Arduino “sketch” but rewrote it “my way”, adding a protection to prevent the transmitter from transmitting for more than one second continuously if the noise level increases.
The Arduino part is definitely not needed. However, it makes it very easy to play with the software to experiment with various timings, trigger values, etc.
Should I decide after experimenting from summits that I want to keep using this idea, I would probably replace the Arduino by a couple of 555’s and make the whole circuit small and light enough to fit in a proper SOTA CW go-kit.
Now for the speech-to-morse wish, there are plenty of speech recognition tools that can generate text (like subtitles) from speech, and also lots of programs that convert text to morse code. So it would definitely be technically feasible, but probably less reliable than using a keyboard to generate the morse code directly.
Probably not, given all the other hindrances and obstacles present in the average SOTA activation. But maybe in 10 (?) years from now, when the newest AI of the time could be squeezed onto a chip (Moore’s law) it might be feasible. But probably still not so much fun as interacting with other stations by voice, with all its’ attendant “features.”
)