This is a very good question so I’ll try my best to answer:
I tried to make my letter pair image system based upon the sound of the first syllable. At first, I was making things up on the fly and writing down ones that I liked and searching for images that I did not have anything good for. I decided that I was not going to stretch the phonetics at all unless I needed to and for most pairs like CE=cement and IC=ice, I was fine.
I did end up stretching what some of those sounds meant though. For example, VF=verf and VW=vow. Those may make absolutely no sense, but I made them into good images by stretching it a bit. Verfassung means constitution in German so I imagine Nicholas Cage from national treasure with the declaration of independence. That was a large stretch, but I automatically think of that. Vow was a character from Wedding Crashers because there was a lot of “Vowing” going on. QM=“Qualm” which is German for a dark heavy smoke
The sounds were more important than the actual letters for some like IF=eiffel tower because it sounds like IF and BQ=bok choy. I would rather use a single syllable BoK instead of an acronym like BBQ because of the phonetics.
There were 2 cases that I broke my Syllabic rules almost completely. IQ and OQ. I made IQ= Albert Einstein because he has a high IQ and OQ was “Officer Qualities” and I imagined someone from the “canadian grenadier guard” because they have cool hats. Since these were the only two that were not following the system I just know them by heart. They still come naturally. Since I use sounds for short term memory I would use “Ick” like the word icky for the sound of IQ and I use the sound “OuaQ” for OQ. I can not mix up ick with the sound and image I use for IK which is “ike”. They clearly sound different and that is good. Over time it all means the same thing and Ouak is clearly now my made up word for a member of the Canadian guard.
This last type of stretching is the type that I was most reluctant about. Some letter pairs between two vowels do not make one syllable sounds. Some do like EI=Eisenhower and EA=Eagle, but what about AA, EE and EO? I used an AA battery, and EEG(Electroencephalography) and Eowyn. I don’t like that they are not single syllables but I actually remember these very well because I can remember during recall if I used an exception like this.
I use the actual letters for single letters, but I put time into making them distinct. For example, I imagine the letter S as a light translucent green letter with a sharp convex 3D shape and some Seaweed hanging from it. These kinds of details kind of build up over the time that I put into my images when I am not memorizing.
That sums up all of the stretching that I did to my images. As long as the sounds were ok, then I didn’t do any of this. I guess some of the ones that have good phonetics were exaggerated a bit as well. For example, I normally imagine a very large Bass and Squirrel. I don’t know why either, I just do that naturally. I might have been trying to add some personality to the squirrel at some point by making him capable of eating people, but I forget if that was planed or not.
That may look bad, but for a 702 image system, I don’t think that that was too much stretching considering the fact that I wrote most of the stretches above. I admit that finding good words would have been more difficult without the internet. It is a very good tool for this and I did use scramble word finders and google searches like “words starting with z containing i” to help construct make my list.
I have not attempted a system as large as yours, but I hope this can still help.
Good Luck!
