Brainstorming a journal app, which may already exist
- text input or audio recording
- whisper.cpp transcription
- user-selected, local AI model that can help navigate obstacles, reflect, or search for content
- offline support
- private, open source
The immediate challenges I see are that transcription and local AI are memory intensive and aren’t practical to live only on frontend so it may not be possible as a PWA without compromise
Any other ideas/thoughts on this?
Not sure if nostr fits in since it is intended to be entirely personal and private
nostr:nprofile1qqs04xzt6ldm9qhs0ctw0t58kf4z57umjzmjg6jywu0seadwtqqc75s8fsrrg nostr:nprofile1qqs99d9qw67th0wr5xh05de4s9k0wjvnkxudkgptq8yg83vtulad30gxyk5sf
#asknostr
I would love something like this! And surely running local, I don't want my crazy thoughts to be wandering in the wilderness of internetz.
Languages may be a hurdle. Some people prefer to think in their fringe mother tongue and transcriptions may be less good in those. Also whenever I try AI in such a language, the responses are not perfect.(AI seems to throw in occasionally some Croatian into Czech or Slovak).
In your experience, do the built-in dictation features work well for the different languages? I don’t know if you’re on iOS or Android. Curious if they handle them well, given that they’ve been refining these speech-to-text algorithms for decades.
Last time I tried Google and Futo, none of them had Czech or Slovak. But I also use GrapheneOS which may be limiting my options.
