Ask HN: Can you train yourself into having good ideas?

I have the technical capabilities to build almost anything I’d want to build, and am generally successful at building things for other people, BUT- my own ideas for projects tend to be atrociously out-of-touch with anything any normal person would actually want/benefit from.

Of course there’s no silver bullet, but is there any way you’ve found to systematically figure out what people’s actual problems are? (for which tech may be a solution, maybe that they’d pay for)

Or is it a sheer volume of output/trial-and-error type of game?