Hey HN, I'm Matt. For the past couple of weeks, I've been hacking on Freestyle. It's an open source voice dictation app.
I was an avid Wispr Flow user before but didn't like how it was cloud only. Every thought sent to their servers for transcription.
I've been working on an alternative that lets you run models locally, like Whisper, Qwen, Sensevoice. That way thoughts never leave your device. Would really appreciate any feedback!!