I talk most of my AI-coding prompts now instead of typing them, and I wanted a tool that does more than transcribe: you speak, and it gives you finished text where your cursor is — a tidy paragraph, a commit message, a ticket — using a small model that runs locally.
That's BoloMic. Whisper does the transcription on-device, a local LLM does the cleanup and the "make this a commit message" step, and it pastes into whatever app you're in. Nothing touches the network unless you choose to plug in your own cloud key.
The reason I'm building it local-first: the tools that already do the smart "finished text" part are cloud-based — they upload your audio, some screenshot your active window — so they're a hard no on any machine touching proprietary or regulated code. BoloMic is for the people that locks out.
I'm at the validation stage, not the "buy now" stage. Genuinely want to know:
Do you hit the "I can't use the cloud dictation tool at work" wall?
Would you trust a local tool more if the engine were open source?
Which would you pay for — a one-time price, or a small flat fee — over yet another $15/mo subscription?
Concept + waitlist if you want to follow along: bolomic.com. Happy to answer anything. (Mac first; Windows on the roadmap — there's a platform picker on the page if you want to vote Windows.)