AI Features Guide

Quick Start

Enable AI Features

1. Open Text Turbo Settings (click extension icon → gear icon)
2. Scroll to AI Settings section
3. Toggle Enable AI Features ON
4. Accept the consent dialog (model will download)
5. Wait for download to complete (~2GB)

Keyboard Shortcuts

Note: On Windows, use Ctrl instead of Cmd

AI Providers

Gemini Nano (Recommended)

Chrome's built-in AI that runs 100% locally on your device.

Requirements:

• Chrome 127 or later
• ~2GB free disk space
• Pro or Trial license

Advantages:

• Completely free
• No internet required (after download)
• Maximum privacy
• Fast responses

Ollama (Advanced)

Run larger AI models locally on your machine for better quality responses.

Requirements:

• Ollama installed (ollama.com)
• At least one model downloaded

Setup:

1. Install Ollama from ollama.com
2. Download a model: ollama pull llama3.2
3. In Text Turbo Settings, select Ollama as provider
4. Enter server URL (default: http://localhost:11434)

Recommended models:

• llama3.2 - Fast, good for general tasks
• mistral - Excellent for writing
• codellama - Best for code-related tasks

AI Actions

Quick Actions (Context Menu)

Right-click on selected text to see AI actions:

• Improve - Enhance clarity and readability
• Grammar - Fix grammar and spelling
• Casual - Make text more conversational
• Formal - Make text more professional
• Expand - Add more detail
• Summarize - Condense text
• Translate - Translate to another language

Magic AI (Shift+Cmd+Enter)

The fastest way to transform text:

1. Type a command in any text field (e.g., "random email", "5 user names")
2. Press Shift+Cmd+Enter
3. AI instantly replaces your text with the result

Works everywhere: forms, chats, code editors, email clients.

Floating Icon Menu (Shift+Cmd+K)

1. Select text on any page
2. Press Shift+Cmd+K or click the floating icon
3. Choose an action from the radial menu

AI Chat

Open AI Chat with Shift+Cmd+A or click the AI tab in Content Modal.

Features:

• Persistent chat history
• Context-aware responses
• Copy responses to clipboard
• Export conversations
• Tab-specific context isolation
• LaTeX/math formula rendering

Tips:

• Ask follow-up questions naturally
• Use clear, specific prompts
• Chat history saves automatically

Custom AI Prompts

Create your own AI actions with custom prompts.

Create a Custom Prompt

1. Go to Settings → AI Settings → Custom Prompts
2. Click "Add Custom Prompt"
3. Enter:
   • Name: Action name (e.g., "LinkedIn Style")
   • Prompt: Instructions for AI (e.g., "Rewrite this text for LinkedIn post")
   • Icon: Choose from Font Awesome or upload custom icon
4. Click Save

Use Custom Prompts

Your custom prompts appear in:

• Right-click context menu
• Floating icon radial menu
• AI Actions Menu

Voice Input

Dictate text instead of typing.

How to Use

1. Click in any text field
2. Double-tap Shift key quickly
3. Speak your text
4. Press Escape or double-tap Shift again to stop

Features:

• Works in any text field (input, textarea, contentEditable)
• Real-time text insertion as you speak
• Uses your extension language setting (20+ languages supported)
• Preserves existing text in field

Tips:

• Speak clearly at normal pace
• Change language in Settings → Language
• Say punctuation if needed ("comma", "period")

Privacy & Security

Local Processing

• Gemini Nano: Runs entirely in Chrome, no data leaves your browser
• Ollama: Runs on your local machine, no cloud transmission

PII Detection

Text Turbo can detect sensitive personal information (SSN, credit cards, emails, phone numbers) before sending to AI.

Detection Modes:

• All providers (recommended): Warns even for local AI (Gemini Nano, Ollama)
• External APIs only: Only warns when using cloud providers (Claude, OpenAI)
• Disabled: No PII detection

Configure:

Settings → AI Settings → Privacy Settings → PII Detection Mode

Site Blacklist

AI is automatically blocked on sensitive sites (banking, healthcare, government). You can configure whitelist in settings.

Data Storage

• All AI chat history stored locally in your browser
• Custom prompts saved locally
• No data sent to external servers (except Voice Input which uses Google Speech API)

Troubleshooting

Gemini Nano not available

1. Check Chrome version (need 127+) in chrome://version
2. Make sure AI is enabled in Text Turbo Settings
3. Check download status in chrome://components
4. Look for "Optimization Guide On Device Model"
5. Restart Chrome after model downloads

Ollama not connecting

1. Verify Ollama is running: ollama serve
2. Check URL in settings (default: http://localhost:11434)
3. Test in browser: http://localhost:11434/api/tags

AI actions not appearing in context menu

1. Check AI is enabled in Settings
2. Verify "Show default actions" is ON
3. Reload the page

Slow responses

• Gemini Nano: First response may be slower (model loading)
• Ollama: Use smaller models for faster responses

Voice Input not working

1. Check microphone permissions in browser
2. Verify internet connection
3. Make sure Voice Input is enabled in Settings → AI

FAQ

Need Help?

• Settings: Click extension icon → gear icon → AI Settings
• Support: support@text-turbo.com
• Demo: Interactive Demo Tour