@oblomov @mckra1g @artemis As a matter of fact LLMs can do that if you instruct them to do so, likewise it can interpret made up words rather well.