twitter.com/disquiet: Voice Recognition, Robot Dreams

I do this manually each week, collating tweets I made at twitter.com/disquiet, my public notebook. Some tweets pop up (in expanded form) on Disquiet.com sooner. It’s personally informative to revisit the previous week of thinking out loud.

▰ I didn’t tweet that much this week, but I did ride my bike for the first time in a long time, so I may live longer and tweet less per week, yielding the same number of tweets in the end as if I’d lived shorter and tweeted more. Or something like that.

▰ Turning my phone’s alarm off isn’t easy, yet once in a while I manage to do it with no memory of having done it. Today I woke a half hour late when a robot called to renew some insurance policy I don’t actually have, which in the past would itself have been a dream narrative.

▰ These not uncommon words seem to be black diamond ski slopes for voice-recognition systems:

“midst”
“amidst”

After I mentioned this, a friend asked why didn’t I just say “amid.” It’s a valid question.

▰ I’ve never actually played Cyberpunk 2077, but I’ve spent an enormous amount of time with YouTube videos of its ambient street noise playing on loop.

▰ These are among the quotidian sounds being turned into music so far in the week’s Disquiet Junto project (disquiet.com/0487). The project was inspired by the carillon, which some 500 years back “instrumentalized” the common bell by letting you plan them, each bell with a different tune.

staircase creaking

elevator recording “ruined” by RF interference

coffee grinder

spring peepers (frogs)

door leading to a small room

▰ And on that note, have a great weekend.