Voice Control (Universal Mic)

Zygotrix AI features a powerful voice control system that lets you interact with the application using natural speech. The Universal Mic provides hands-free navigation, voice commands, and dictation capabilities.

Overview

The Universal Mic is always visible in the header as a magic wand icon. It provides:

• Voice Commands - Navigate the app, toggle settings, and execute actions
• Dictation Mode - Speak your messages instead of typing
• AI Responses - Get spoken feedback powered by OpenAI
• Customizable Voice - Adjust speed, pitch, and voice character

Getting Started

Activating Voice Control

Click the magic wand button in the header to activate voice control:

• Inactive State: Gray circular button with magic wand icon
• Active State: Emerald green button with animated sparkle

When activated, the AI will greet you with a welcome message like:

"Hi! I'm your voice assistant for Zygotrix. How can I help you today?"

Deactivating Voice Control

Say any of these commands to stop listening:

• "bye" / "bye bye"
• "goodbye"
• "exit" / "quit"
• "close yourself"
• "stop listening"

The AI will respond with a farewell and automatically turn off.

Voice Commands

Navigation Commands

Command	Action
"Go to settings"	Opens the Settings page
"Go to chat"	Opens the AI Chat page
"New chat" / "New conversation"	Starts a new chat
"Open sidebar"	Opens/expands the sidebar
"Close sidebar"	Closes/collapses the sidebar

Utility Commands

Command	Action
"Toggle theme" / "Switch theme"	Switches between light and dark mode
"Show usage"	Opens your usage statistics
"Search for [query]"	Searches your conversations
"Clear search"	Clears the search input

Conversational Commands

Command	Response
"Hi" / "Hello" / "Hey"	Friendly greeting response
"Bye" / "Goodbye" / "Exit"	Farewell and deactivates mic

Unrecognized Commands

If you say something the AI doesn't recognize, it will politely respond:

"I'm sorry, I don't have that ability yet. Would you like me to help with navigation, settings, or something else?"

Dictation Mode

When you focus on the chat input field, voice control switches to dictation mode:

1. Click on the chat input field
2. Start speaking your message
3. Watch your words appear in real-time
4. Press Enter or click Send when done

Visual Indicator: A premium floating pill appears at the top of the screen showing:

• Mic icon with pulse animation
• "Listening..." label
• Live transcript of your speech
• Sound wave visualizer

Voice Settings

Customize your AI assistant's voice in Settings → Voice:

Voice Character

Choose from available system voices. Each voice has a unique personality:

• Different accents (US, UK, Australian, etc.)
• Male and female options
• Multiple languages supported

Speaking Speed (Rate)

Adjust how fast the AI speaks:

• Slow (0.5x) - For careful listening
• Normal (1.0x) - Standard speed
• Default (1.2x) - Slightly faster, natural flow
• Fast (2.0x) - Quick responses

Voice Pitch

Adjust the tone of the voice:

• Low (0.5) - Deeper voice
• Normal (1.0) - Standard pitch
• High (2.0) - Higher voice

Preview Button

Click "Preview Voice" to hear a sample with your current settings:

"Hello! This is how I sound with your current settings. I'm your Zygotrix voice assistant."

Reset to Defaults

One-click reset returns settings to:

• Rate: 1.2x
• Pitch: 1.0
• Voice: First available

Technical Details

How It Works

1. Speech Recognition: Uses the Web Speech API (SpeechRecognition) to convert speech to text
2. Intent Analysis: OpenAI GPT-4o-mini analyzes the transcript and matches it to available commands
3. Command Execution: The matched command is executed
4. Voice Response: Text-to-Speech (speechSynthesis) speaks the response

AI-Powered Responses

Voice responses are generated dynamically by AI, not hardcoded:

• Welcome Message: Generated when mic activates
• Fallback Response: Generated when no command matches
• Error Handling: Spoken feedback on errors

Browser Compatibility

Browser	Speech Recognition	Text-to-Speech
Chrome	Full support	Full support
Edge	Full support	Full support
Safari	Partial	Full support
Firefox	Not supported	Full support

tip

For the best experience, use Google Chrome or Microsoft Edge.

Permissions

Voice control requires microphone permission:

1. Click the mic button
2. Browser will prompt for permission
3. Click "Allow" to enable voice control

Troubleshooting

"Voice control not working"

1. Check that your browser supports Speech Recognition (Chrome/Edge recommended)
2. Ensure microphone permission is granted
3. Check that your microphone is not muted at the system level

"Commands not recognized"

• Speak clearly and at a normal pace
• Wait for the speech to be processed before speaking again
• Try rephrasing your command

"No audio response"

1. Check your system volume
2. Go to Settings → Voice and click "Preview Voice"
3. Try selecting a different voice character

"Wrong voice selected"

1. Go to Settings → Voice
2. Select a different voice from the dropdown
3. Click "Preview Voice" to test
4. Settings are saved automatically

Best Practices

For Voice Commands

Do: Use clear, direct commands

"Go to settings"

Don't: Use complex sentences

"I would like to maybe go to the settings page if possible"

For Dictation

Do: Speak in complete thoughts, pause naturally

"What is the result of crossing Aa with Aa"

Don't: Speak too fast or mumble

"whatistheresultofcrossingaawithaa"

For Best Recognition

• Minimize background noise
• Speak at a moderate pace
• Use a quality microphone
• Keep reasonable distance from mic

Privacy

• Voice data is processed locally by your browser's Speech Recognition API
• Transcripts are sent to OpenAI for intent analysis
• No audio recordings are stored
• Voice settings are saved locally in your browser

API Reference

For developers, the Voice Control system is exposed via React Context:

import { useVoiceControl } from "../contexts";

const MyComponent = () => {
  const {
    isListening, // Boolean: is mic active
    toggleListening, // Function: toggle mic on/off
    speak, // Function: speak text
    registerCommand, // Function: add custom command
    voiceSettings, // Object: current settings
    setVoiceSettings, // Function: update settings
  } = useVoiceControl();

  return (
    <button onClick={toggleListening}>
      {isListening ? "Stop" : "Start"} Listening
    </button>
  );
};

Registering Custom Commands

useEffect(() => {
  const unregister = registerCommand(
    "my-command", // Unique ID
    (text) => {
      // Action function
      console.log("Command triggered!", text);
      speak("Executing my command");
    },
    "Description for AI to match against"
  );

  return unregister; // Cleanup on unmount
}, [registerCommand, speak]);