Blog

Resonant's MCP server turns your voice workspace into a knowledge base for any AI agent, including Claude and Codex. Your meetings, dictations, and memos — queryable by your AI tools.

NVIDIA Parakeet v3 and Qwen3 ASR, both compiled to CoreML and running on Apple Neural Engine. A technical look at how Resonant does on-device speech recognition.

Dictating in VS Code, Slack, and Gmail should produce different text. Resonant captures the app, window, URL, and selected text — and adapts its output to match.

AI cleanup should fix your grammar, not rewrite your meaning. Resonant's cloud processing has guardrails that detect when an LLM rewrites your dictation instead of cleaning it.

Ambient context records app names, window titles, URLs, and dwell time — not screenshots, keystrokes, or file contents. A transparent breakdown of what Resonant sees.

Watch Resonant transform voice into text instantly. A quick demo of local-first dictation on your Mac.

Cloud dictation works. But it sends every word you say to someone else's server. We think there's a better way — one where your voice never leaves your Mac.

The first build of Resonant is here. It's functional, it's fast, and it's free.