Voice-First Input for AI Coding

Hold a hotkey, speak, and your words become an engineering prompt — transcribed locally, refined by an LLM, injected into your AI dev tool. Built entirely with AI.

🎙️ Push-to-Talk Local transcription · LLM refinement · One-tap injection

Swift 6 macOS 14+ WhisperKit 100% AI Built

How It Works

Screenshots and a demo video are coming soon. The chain below is already working end-to-end.

🎙️

1. Hold & Speak

Hold the global hotkey (default Right Option) and speak naturally. A 200 ms hold-to-confirm guard prevents mistouches.

🔒

2. Transcribe Locally

Release to transcribe on-device with WhisperKit small — audio stays in memory and is never written to disk.

3. Refine

Your own LLM key turns the transcript into a clean engineering prompt, with hard anti-hallucination constraints.

📋

4. Inject

The result lands at the cursor of the app you were using — clipboard + Cmd+V with restore and race protection.

🤖 100% AI Built

This project was entirely built with AI coding assistants. Every line of code was generated, reviewed, and refined through AI collaboration.

Kimi Code CLI

Kimi K3

DeepSeek

V4 Pro

Features

🎙️

Push-to-Talk Hotkey

Default Right Option with 200 ms confirmation, 300 ms mistouch cancel and Esc to discard. Presets: Right Option / Right Command / Right Shift / Fn.

🔒

Local, Private Transcription

WhisperKit small runs fully on-device with Chinese/English mixed support; Speech framework fallback until the model is ready.

LLM Prompt Refinement

Removes filler words, normalizes phrasing, resolves anaphora from context, and falls back to raw text on timeout — bring your own key.

🎯

Context-Aware Injection

Snapshots the frontmost app and window title on press, re-checks on release, and classifies terminal / editor / browser / other.

📋

Clipboard + Cmd+V

Clipboard snapshot/restore with change-count race protection; optional auto-send (Return); newline folding for CLI targets.

🪟

Non-Activating HUD

Waveform, stage status and success/failure feedback — visible in full screen and across Spaces without stealing focus.

🛡️

Permission Self-Check

Guided onboarding with deep links into System Settings and graceful degradation for mic / input monitoring / accessibility.

🩺

Diagnostics

Categorized os_log plus on-disk logs with one-click export from Settings.

Tech Stack

Swift 6 SwiftUI AppKit AVAudioEngine CGEventTap WhisperKit Speech Keychain Swift Testing