🔊 Speech Synthesis Voices

Detect installed text-to-speech (TTS) voices in your browser. Test voices interactively, explore voice fingerprinting, and understand how voices reveal your OS and language packs.

Total Voices

Local Voices

Network Voices

Languages

Default Locale

Test Speech Synthesis

Select a voice and click "Speak" to test how it sounds

Voice Fingerprint Hash

Calculating...

This hash uniquely identifies your voice configuration (click to copy)

Privacy & Fingerprinting

Speech synthesis voices are a powerful fingerprinting vector. Different operating systems and language packs have unique voice sets. Combined with other browser properties, your voice configuration can create a unique identifier for tracking.

The Web Speech API provides speech synthesis (text-to-speech) capabilities to web applications. The window.speechSynthesis interface allows JavaScript to:

List available voices - Get all TTS voices installed on the system
Speak text - Convert text to speech using selected voice
Control speech - Pause, resume, cancel speech
Customize voice - Adjust rate, pitch, volume

Voice Properties

voiceURI - Unique identifier for the voice
name - Human-readable voice name
lang - BCP 47 language tag (e.g., "en-US", "fr-FR")
localService - true if local, false if network/cloud
default - true if this is the default voice for its language

// Get all available voices
const voices = window.speechSynthesis.getVoices();
console.log(`Found ${voices.length} voices`);

voices.forEach(voice => {
    console.log(`${voice.name} (${voice.lang})`);
    console.log(`  Local: ${voice.localService}`);
    console.log(`  Default: ${voice.default}`);
});
                

Speech synthesis voices create a unique fingerprint because:

1. Operating System Detection

Windows - Microsoft David, Zira, Mark voices
macOS - Alex, Samantha, Victoria voices
Linux - eSpeak voices, or none if TTS not installed
iOS - Siri voices in multiple languages
Android - Google TTS voices

2. Language Pack Detection

Installed voices reveal which language packs the user has installed. For example, if Japanese, Arabic, and Hindi voices are present, it suggests the user works with multiple languages or is from a multilingual region.

Special case: Russian voices are particularly revealing. Russian TTS voices are typically only present on systems with Russian localization or Russian language packs explicitly installed. This makes Russian voices a strong fingerprinting signal, as they indicate either a Russian-speaking user or someone who works extensively with Russian content.

3. System Customization

Users who install additional voice packs (like premium voices or language-specific TTS) create more unique fingerprints. The combination of voices is highly distinctive.

4. Voice Count & Combinations

The number of voices and their specific combinations create thousands of possible fingerprint values. Combined with other browser properties, this makes tracking very effective.

// Example: Create voice fingerprint hash
function generateVoiceFingerprint() {
    const voices = window.speechSynthesis.getVoices();

    const voiceSignature = voices
        .map(v => `${v.name}|${v.lang}|${v.localService}`)
        .sort()
        .join(',');

    // Simple hash function (SHA-256 recommended for production)
    let hash = 0;
    for (let i = 0; i < voiceSignature.length; i++) {
        const char = voiceSignature.charCodeAt(i);
        hash = ((hash << 5) - hash) + char;
        hash = hash & hash;
    }

    return Math.abs(hash).toString(16).padStart(8, '0');
}

console.log('Voice Fingerprint:', generateVoiceFingerprint());
                

Here's how to use the Speech Synthesis API in your own applications:

Basic Text-to-Speech

// Create an utterance
const utterance = new SpeechSynthesisUtterance('Hello, world!');

// Speak the text
window.speechSynthesis.speak(utterance);
                

Select Specific Voice

// Get voices
const voices = window.speechSynthesis.getVoices();

// Find a specific voice
const selectedVoice = voices.find(v =>
    v.name === 'Google US English' || v.lang === 'en-US'
);

// Create utterance with selected voice
const utterance = new SpeechSynthesisUtterance('Testing voice selection');
utterance.voice = selectedVoice;
utterance.rate = 1.0;    // Speed (0.1 to 10)
utterance.pitch = 1.0;   // Pitch (0 to 2)
utterance.volume = 1.0;  // Volume (0 to 1)

window.speechSynthesis.speak(utterance);
                

Event Listeners

const utterance = new SpeechSynthesisUtterance('Hello!');

utterance.addEventListener('start', () => {
    console.log('Speech started');
});

utterance.addEventListener('end', () => {
    console.log('Speech ended');
});

utterance.addEventListener('pause', () => {
    console.log('Speech paused');
});

utterance.addEventListener('error', (event) => {
    console.error('Speech error:', event.error);
});

window.speechSynthesis.speak(utterance);
                

Control Speech

// Pause speech
window.speechSynthesis.pause();

// Resume speech
window.speechSynthesis.resume();

// Cancel all speech
window.speechSynthesis.cancel();

// Check if speaking
const isSpeaking = window.speechSynthesis.speaking;
console.log('Currently speaking:', isSpeaking);
                