diff --git a/.github/workflows/pages.yml b/.github/workflows/pages.yml new file mode 100644 index 00000000..65dad439 --- /dev/null +++ b/.github/workflows/pages.yml @@ -0,0 +1,31 @@ +name: Deploy GitHub Pages + +on: + push: + branches: [main] + paths: [docs/**] + workflow_dispatch: + +permissions: + contents: read + pages: write + id-token: write + +concurrency: + group: pages + cancel-in-progress: false + +jobs: + deploy: + environment: + name: github-pages + url: ${{ steps.deployment.outputs.page_url }} + runs-on: ubuntu-latest + steps: + - uses: actions/checkout@v4 + - uses: actions/configure-pages@v5 + - uses: actions/upload-pages-artifact@v3 + with: + path: docs + - id: deployment + uses: actions/deploy-pages@v4 diff --git a/Docs/index.html b/Docs/index.html new file mode 100644 index 00000000..72e6fd29 --- /dev/null +++ b/Docs/index.html @@ -0,0 +1,1103 @@ + + + + + + OpenType — AI Voice Input for macOS + + + + + + + + + + + + + + + +
+
+
Now Available · macOS Tahoe
+

Speak.
Think.
Type.

+

+ AI-powered voice input that lives in your menu bar. + Hold a key, speak naturally — your words appear instantly, polished and precise. +

+
+ +
+
+ + + Download .dmg + + + + View Source + +
+
+
Platform macOS 26+
+
Chip Apple Silicon
+
Language Swift 6
+
License MIT
+
+
+ + +
+
+ Menu Bar Popover + OpenType menu bar popover interface +
+
+ + +
+ How It Works +

Three steps.
Zero friction.

+
+
+
01
+
Press & Hold
+

Hold your configured key — Fn, Ctrl, Shift, or Option. Recording begins instantly. A subtle overlay confirms you're live.

+
⌘ fn
+
+
+
02
+
Speak Naturally
+

Talk as you would to a person. WhisperKit runs entirely on your device — no audio ever leaves your Mac.

+
on-device
+
+
+
03
+
Release & Insert
+

Release the key. Your speech is transcribed, optionally refined by a local LLM, and inserted directly into your active app.

+
any app
+
+
+
04
+
Context-Aware
+

Screen OCR captures what's visible. The AI uses this to fix homophones and align output with your actual context.

+
ocr + vision
+
+
+
+ + +
+ Output Modes +

Three ways to speak.

+

Pick the mode that matches your task — from raw dictation to AI-powered command execution.

+
+
+
+
⚡
+
Mode 01
+
Verbatim
+

Raw transcription with the lowest possible latency. What you say is exactly what gets typed.

+ Lowest Latency +
+
+
✦
+
Mode 02
+
Smart Format
+

An LLM cleans up filler words, fixes grammar, and structures your speech into polished text.

+ AI-Refined +
+
+
+
+
◈
+
Mode 03
+
Voice Command
+

Speak a command. The AI reads your screen context via OCR and generates a contextually relevant response — summarize, translate, reply.

+ Screen-Aware + AI-Generated +
+
+

+ "All three modes use your personal Edit Rules and Language Style Presets — concise, formal, casual, or fully custom prompts." +

+
+
+
+
+ + +
+
+
+ Features +

Everything you need.
Nothing you don't.

+
+

Built for power users who care about privacy, performance, and control over their tools.

+
+
+
+
🎙
+
Triple Speech Engines
+

Apple Speech (built-in), WhisperKit (offline Whisper), or Doubao ASR (cloud). Pick what fits your workflow.

+
+
+
🔒
+
Fully Local Inference
+

WhisperKit + MLX Qwen runs entirely on-device. Audio and text never leave your machine.

+
+
+
🧠
+
Local LLM (Qwen3)
+

MLX-powered Qwen2.5 and Qwen3 models for on-device smart formatting and correction.

+
+
+
🌐
+
Remote LLM Support
+

OpenAI, Claude, Gemini, OpenRouter, SiliconFlow, Doubao, Bailian, MiniMax — your key, your model.

+
+
+
⌨️
+
Global Hotkey
+

Configure any modifier key with long-press, double-tap, or single-tap trigger modes.

+
+
+
👁
+
Screen OCR Context
+

ScreenCaptureKit + Vision captures on-screen text to help the LLM correct homophones in context.

+
+
+
📜
+
Input Memory
+

Recent inputs are injected as LLM context, improving accuracy for continuous dictation sessions.

+
+
+
✏️
+
Edit Rules
+

Personal text replacement rules applied automatically to every output. Your voice, your vocabulary.

+
+
+
📊
+
Input History & Stats
+

Full history with raw vs. processed comparison, word counts, and configurable retention period.

+
+
+
+ + +
+ Tech Stack +

Built on the best of
Apple Silicon.

+
+
+
+
WhisperKit
+
Offline speech recognition
+
+
+
+
MLX-Swift-LM
+
Local LLM on Apple Silicon
+
+
+
+
Swift 6
+
Core language
+
+
+
+
SwiftUI + AppKit
+
Native macOS UI
+
+
+
+
ScreenCaptureKit
+
Screen context capture
+
+
+
+
AVAudioEngine
+
Low-latency mic capture
+
+
+
+
Vision Framework
+
On-device OCR
+
+
+
+
Apple Speech
+
Built-in ASR engine
+
+
+
+ + +
+ Remote LLM Providers +

Works with every
major AI provider.

+
+
OpenAI
+
Anthropic Claude
+
Google Gemini
+
OpenRouter
+
SiliconFlow
+
Volcengine Doubao
+
Alibaba Bailian
+
MiniMax China
+
MiniMax Global
+
+ Any OpenAI-compatible API
+
+
+ + +
+

Ready to
stop typing?

+

Download the latest release. Drag to Applications. Hold Fn. Start speaking.

+
+ + + Download for macOS + + + + View on GitHub + +
+
+ xattr -cr /Applications/OpenType.app + +
+

Run this once after install if macOS shows a security prompt

+
+ + + + + + +