Hey Product Hunt! 👋

I'm the maker of RocketWhisper — a native macOS app that handles both ends of voice, 100% offline: talk into any Mac app, OR let your Mac listen and translate any video/meeting/podcast for you.

Why I built this

I'm a developer who dictates more than I type. Existing voice-to-text tools either (1) send audio to the cloud, (2) require an API key and per-minute fees, or (3) ship Electron and eat 500MB of RAM. None of them felt right on a Mac.

When Apple shipped SpeechAnalyzer, Foundation Models, and the Translation framework in macOS 26, I realized the whole stack — recognition, formatting, translation — could finally run on-device. Fast, private, free of API keys.

What makes it different

🎙️ Dictate into ANY Mac app via ⌥Space

🎬 Live Translation — Netflix-style floating subtitles over any video/meeting/podcast (40+ languages)

🧠 Apple Intelligence native: SpeechAnalyzer + Foundation Models + Translation framework on macOS 26 (WhisperKit fallback on macOS 14-15)

⚡ Optimized for MacBook Neo (A18 Pro / 8GB RAM) but runs on any Apple Silicon Mac

🔒 Fully offline — your voice never leaves your Mac

🌍 Full English + Japanese support

💰 One-time purchase, no subscription

Also works with WhisperKit + OpenAI / Claude / Gemini / LM Studio if you prefer those.

What I'd love feedback on

- Which language pair would you use Live Translation for most?

- How does the voice command system feel (e.g. "new line", "new paragraph")?

- Anything you'd want in Stage 4 LLM formatting?

- Pricing — does a one-time purchase model resonate?

Happy to answer anything. Thanks for checking it out! 🚀