Here is a simple example that will write the current audio stream (in this case a soundfile) to a table, that is then used by the synth instrument that plays on a note press. Note that I also apply a fade to the table as otherwise the discontinuities will cause clipping. Also note that I don't have speakers to really test this on right now!
fileToWavetable.csd (1.5 KB)
You will need to swap out the sound file for you own. Of course you will want to add lots of extra options such as the ability to load any sound file. But note that this brute force approach may lead to lots of aliasing.
Dumping the table to a txt file should be simply enough. As you say, GEN23 can then easily read back the values. Sorry I don't have time to prepare a better example, but this should get you moving in the right direction at least