You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
I saw under the TODO the ability generate random words, and in that vein, it would be cool to be able to generate Geordi-style technobabble. Like 'If we could just recalibrate the plasma flux induction matrix and reroute the output through oscillating magnetocoil, we might just be able to get enough throughput on the phase inverter.'
The text was updated successfully, but these errors were encountered:
Haha, that sounds like it would be a fun one, but certainly introduces an NLP challenge where there wasn't one before. Are you suggesting that we create new sentences from a hardcoded set of technobabble (à la Chomskybot), or identify lines from TNG that are already technobabbly and have a flag in the generator that makes it choose only those?
My thinking was since Geordi's and Data's lines have a high percentage of technobabble, it would be human-manageable to pick out a good sample of words, and then split them up into a couple of types and then have some formats like:
'Perhaps we could the <general babble 1> <general babble 2>'
'If we could just the <general babble 1> <general babble 2>, then we might be able to ...'
The nice thing is that since much of his technobabble is hypothetical, we could get away with infinitives for verbs, and the nouns are pretty generalized, too. I'm going to try to take a pass at this and see what I come up with. NLP's not really my thing, but I think a rough generator would be pretty manageable.
I saw under the
TODO
the ability generate random words, and in that vein, it would be cool to be able to generate Geordi-style technobabble. Like 'If we could just recalibrate the plasma flux induction matrix and reroute the output through oscillating magnetocoil, we might just be able to get enough throughput on the phase inverter.'The text was updated successfully, but these errors were encountered: