In my experience all the best free software comes from a developer scratching his/her own personal itch. It's really unlikely, on a planet with several billion people, that you're the only person with any given itch to scratch, and the beauty of free software is other people don't have to overcome any barriers to participating in scratching your itch, save the unavoidable ones of capability and desire.
Look at the feed of new and updated software on https://f-droid.org/ for examples. There's (inter alia) an app for talking to the home-automation dashboard HABPanel, two language keyboard packs (Afrikaans and Galician), a J2ME emulator, one that monitors the status of the Stratum 0 Hackerspace and can open the door with your ssh key, and a mini space battle. Other gems include one to read Swedish and Danish public travelcards via NFC, one to find vegetarian and vegan restaurants, a trivia challenge about the Apollo moon missions, and one to easily mange the wireless network at Chemnitz University of Technology.
None of these feels to me like the result of a research process, then a slow identification of people who might be interested, a design process, then some coding and review, followed by UA testing, leading to release. They feel like someone got up one morning and went "hey, I'm fed up not being able to (easily) do $WEIRD_THING
, I can't find any existing software to help me do it, so I'm going to write some".
Just get on with writing software to do something that you need done; people, feedback, ideas, and help will likely come, and if they don't, you'll still have one less annoying un-automated task in your life.