Incognito PrivoAI

Free, no backend · data in browser. Voice-to-text, AI chat, web translation & AI voice reading. Your Gemini API key — full privacy.

🆕 v27 & v28 Highlights (2026: v27.2 → v28) ──────────────────────────────────────────── ▶ 2026.3 – Google released previews of Gemini 3.1 Pro and 3.1 Flash Lite. The workbench keeps the stable 2.5 Flash; 3.1 Flash Lite is now the default, and 3.1 Pro is available in chat. (v28.0) ▶ Celebrating the 2026 World Baseball Classic Group C matchups — Czech language support added. (v28.0.1) ▶ Rebranded as Incognito PrivoAI, combining Private, Voice & AI. ▶ 2026.2 – Local AI model support added (v27.2) Image recognition, file parsing, and voice features still rely on Google Cloud / Gemini API. But if your computer is powerful enough, you can connect Ollama or LM Studio for local text chat and translation. ▶ Sort translation target languages (v27.2) Set up to 3 preferred languages in Settings, reorder with ↑↓ arrows — your most-used languages always appear first. ▶ Default prompt presets (v27.2) Incognito Chat offers 7 preset system prompt groups (A–G). Combine AI role, audience, response format, task, and output settings for maximum efficiency. ── Core Features ── 🎙️ Voice Notes Record audio or upload a file — AI converts speech to text and labels different speakers. Tap any paragraph to translate instantly. Perfect for meeting notes, interviews, and multilingual memos. 💬 Incognito AI Chat A private Gemini AI chat window — no history saved after closing. Three Google models available: • Gemini Flash Lite — fastest, great for everyday Q&A and translation • Gemini Flash — balanced speed & quality, supports document analysis • Gemini Pro — strongest reasoning, ideal for deep analysis of long documents All models support image and PDF uploads. 🌐 Translation Workbench Paste text, upload a screenshot, or translate the current webpage. Supports OCR for image text, selectable target language, and AI summary. 📖 Reader & Webpage Translation Converts complex webpages into a clean reading view with translation support. (Recommended as a fallback when Chrome's built-in Google Translate doesn't work.) ▶ Built-in text-to-speech in reader (v27.2) Translated or transcribed text can be read aloud. Select a language and tap "Speak" to play any paragraph. 🔊 AI Voice Reading [Beta] (v28) Convert YouTube captions, audio files, webpages, or pasted text into natural-sounding audio using Gemini TTS (Aoede voice, calm & friendly). Supports custom reading instructions. ── Privacy by Design ── • No backend server — all AI calls go directly from your browser • Zero data collection — we never see your inputs or outputs • Self-managed keys — Gemini API key stored only in Chrome local storage • Local AI support — connect Ollama and keep data entirely on your machine ── Interface Languages ── 13 languages: Traditional Chinese, English, Japanese, Korean, German, Spanish, Arabic (RTL), Russian, Portuguese (Brazil), Turkish, Ukrainian, Vietnamese, Czech. ── Getting Started ── 1. Install the extension 2. Go to Google AI Studio (aistudio.google.com) and get a free Gemini API key 3. Click the extension icon → Settings → paste your key 4. Start using all features — completely free!