Hi, I am in a similar position at a an Indutrial Heritage Museum in Wales UK. I've tried a couple of Morse encode/decode programs/sketches, modified them and tried various things adn in the end decided that to radically simplify my approach. The reasons for this are:
- The average person visiting the museum, with best will in the world, can't send morse with sufficiently precise spacings and mark/space ratios for a reliable decode. They will send, a dot then look up the next dot or dash on the accompanying table and so on. In the meantime the software has given up waiting.
- Off air decode seemed a possibility. I got the software to decode machine generated morse OK. i.e. off computer so it was perfectly sent. I could decode reasonably off the air. The snag is that even if you do decode Amateur transmissions off the air, they contain so many abbreviations and "Q" codes that yet another layer of interpretation is needed as they will not make sense to the average visitor.
- Commercial or military morse is rarely if ever used now, if you do find some it is likely to be abbreviated and probably encrypted. Finally, there willl be copyright issues to contend with.
I decided instead to provide a practice key for people to try their hand and also to program some stock messages in the arduino that could be "sent" and decoded. That way visitors can hear authentic morse and see the decode coming in at the same time. There are some other functions to be included to do with maritime distress calls, but that's another issue. All this will be up and running for the next season starting next Easter (2017).
Hope this helps - Michael.