Hey Product Hunt 👋 Voice is the fastest input we have. We built Dimmy to turn it into a real power tool, on your terms: you choose how it runs and which AI powers it. It's a native app for macOS and Windows with three tools: 🎙️ Dictation: press a hotkey, talk, release. Local Whisper transcribes on-device, free, no account, no API key, and clean text lands wherever your cursor is: editor, browser, chat, terminal. many languages. ⚡ Command mode: select text and say "make this formal" or "turn this into bullets" and it rewrites in place. Or speak a request and it drafts at the cursor. 📝 Meetings: record both sides of a call with no bot joining, and get a structured recap: decisions, action items, open questions, timestamps. Export to Notion, a folder (Obsidian, Drive, Dropbox), or Claude MCP! The part people like: you bring the brain. Command mode and recaps run on the AI you choose, your own API key (Claude, OpenAI, Groq, Gemini), your existing Claude or ChatGPT subscription, or a fully local model if you want zero cloud. So you decide the quality/privacy trade-off, not us. Want it 100% offline? You can. Want a top cloud model? Plug your key, and pay for what you use. It's open-core under AGPL and free if you build from source. The paid plan (€4.99/mo or €99 lifetime) buys a signed, ready-made build and updates, not features. Honest about what it's not: dictation is local and key-free, but command mode and recaps need an AI provider (a key, your subscription, or a local model). No mobile app, yet. We're around all day for questions, and feedback genuinely shapes the roadmap. What would you want your voice to do? 🙏

Dimmy.app

Turn your voice into a superpower! On your terms.

Turn your voice into a superpower! On your terms.