Voice Library - VoxNexus

Overview

VoxNexus offers a diverse library of high-quality voices covering multiple languages, accents, genders, and styles. Each voice is carefully crafted using advanced AI technology to deliver natural, expressive speech.

Language and locale

TTS model languages and each voice locale’s language use BCP 47 language[-Script], for example en, fr, zh-Hans, and zh-Hant. They classify languages and writing systems, not regional accents. Translation uses its separate language code contract. A voice’s supported_locales[].locale uses language[-Script][-REGION], such as en, en-US, or en-GB. Include model_id in voice list, search, and detail requests to obtain the locales actually available for that model and voice. voice_id is a stable opaque identifier; do not infer language or accent from its spelling. Input case and separator aliases are accepted, while responses use canonical tags: EN_us and en-Latn-US become en-US; zh-Hant-TW becomes zh-TW. Explicit nonredundant Script constraints, such as zh-Hant-CN, remain intact. Bare en or fr never gains an invented region. primary_locale, when present, belongs to supported_locales and is the default. A model-filtered voice can omit it when its original default is unavailable and several supported locales remain. Select a locale explicitly in that case. Two English locales such as en-US and en-GB alone do not make a voice multilingual. For synthesis, language=en leaves the region unspecified: the service prefers a matching primary locale, otherwise the only matching locale. Multiple candidates without a default require an explicit selection. language=en-GB requires British English support and cannot silently use en-US. A locale without REGION does not promise a particular regional accent; accent is separate voice metadata. Script describes writing, not accent.

Voice Categories

By Language

Our voice library supports 50+ languages and locales:

English: Multiple accents (US, UK, Australian, etc.)
Chinese: Mandarin, Cantonese, and regional variants
Spanish: European and Latin American variants
French: European and Canadian French
German: Standard German and regional accents
Japanese: Standard Japanese
Korean: Standard Korean
And many more…

By Gender

Choose voices that match your application’s needs:

Male Voices: Deep, authoritative, or friendly male voices
Female Voices: Warm, professional, or energetic female voices
Neutral Voices: Gender-neutral options for inclusive applications

By Category

Voices are organized into categories:

Premade Voices: Professional, pre-built voices ready for immediate use
Cloned Voices: Custom voices created from audio samples
Generated Voices: AI-generated voices with unique characteristics

By Style

Match voice style to your content:

Professional: Formal, business-appropriate voices
Casual: Friendly, conversational voices
Cheerful: Energetic, upbeat voices
Calm: Soothing, relaxed voices
Authoritative: Confident, commanding voices
Friendly: Warm, approachable voices

Discovering Voices

Browse All Voices

Use the /v1/voices endpoint to browse available voices:

curl -X GET "https://api.voxnexus.ai/v1/voices" \
  -H "X-Api-Key: YOUR_API_KEY"

Search Voices

Search for voices by keyword:

curl -X GET "https://api.voxnexus.ai/v1/voices/search?q=xiaoxiao" \
  -H "X-Api-Key: YOUR_API_KEY"

Filter Voices

Filter voices by attributes:

# Filter by language
curl -X GET "https://api.voxnexus.ai/v1/voices?model_id=vn-tts-basic&language=zh-Hans" \
  -H "X-Api-Key: YOUR_API_KEY"

# Filter by locale
curl -X GET "https://api.voxnexus.ai/v1/voices?locale=zh-CN" \
  -H "X-Api-Key: YOUR_API_KEY"

# Filter by gender
curl -X GET "https://api.voxnexus.ai/v1/voices?gender=female" \
  -H "X-Api-Key: YOUR_API_KEY"

# Filter by category
curl -X GET "https://api.voxnexus.ai/v1/voices?category=premade" \
  -H "X-Api-Key: YOUR_API_KEY"

# Filter by style
curl -X GET "https://api.voxnexus.ai/v1/voices?style=cheerful" \
  -H "X-Api-Key: YOUR_API_KEY"

# Filter by model
curl -X GET "https://api.voxnexus.ai/v1/voices?model_id=vn-tts-basic" \
  -H "X-Api-Key: YOUR_API_KEY"

# Combine filters
curl -X GET "https://api.voxnexus.ai/v1/voices?locale=zh-CN&gender=female&style=cheerful" \
  -H "X-Api-Key: YOUR_API_KEY"

Get Voice Details

Retrieve detailed information about a specific voice:

curl -X GET "https://api.voxnexus.ai/v1/voices/vn-xiaoxiao?model_id=vn-tts-basic" \
  -H "X-Api-Key: YOUR_API_KEY"

Voice Response Format

When you query the voice API, you’ll receive responses in this format:

{
  "voices": [
    {
      "voice_id": "vn-xiaoxiao",
      "name": "Xiaoxiao",
      "display_name": "Xiaoxiao",
      "gender": "female",
      "category": "premade",
      "description": "A warm and friendly female voice, suitable for general scenarios",
      "primary_locale": "zh-CN",
      "is_multilingual": false,
      "supported_locales": [
        {
          "locale": "zh-CN",
          "language": "zh-Hans",
          "locale_name": "Chinese (Simplified, China)",
          "accent": "mandarin",
          "preview_url": "https://api.voxnexus.ai/samples/vn-xiaoxiao/zh-CN.mp3",
          "is_primary": true
        }
      ],
      "labels": {
        "personalities": ["warm", "friendly"],
        "scenarios": ["assistant", "news"]
      },
      "preview_url": "https://api.voxnexus.ai/samples/vn-xiaoxiao.mp3",
      "config_schema": {
        "style": {
          "type": "string",
          "enum": ["cheerful", "sad", "angry"]
        },
        "role": {
          "type": "string",
          "enum": ["Girl", "Boy"]
        },
        "style_degree": {
          "type": "number",
          "minimum": 0.01,
          "maximum": 2,
          "default": 1
        }
      },
      "models": ["vn-tts-basic"]
    }
  ],
  "pagination": {
    "page": 1,
    "page_size": 20,
    "total": 150,
    "total_pages": 8
  }
}

Voice Configuration

Many voices support additional configuration options through the voice_config parameter:

Style Configuration

Some voices support different speaking styles:

{
  "text": "Hello, world!",
  "voice_id": "vn-xiaoxiao",
  "voice_config": {
    "style": "cheerful"
  }
}

Role Configuration

Some voices can adopt different roles:

{
  "text": "Hello, world!",
  "voice_id": "vn-xiaoxiao",
  "voice_config": {
    "role": "Girl",
    "style_degree": 0.5
  }
}

Checking Voice Configuration Options

Each voice’s config_schema field describes the available configuration options. Check the voice details to see what’s supported:

const response = await fetch('https://api.voxnexus.ai/v1/voices/vn-xiaoxiao', {
  headers: {
    'X-Api-Key': 'YOUR_API_KEY'
  }
});

const voice = await response.json();
console.log('Available config options:', voice.config_schema);

Popular Voices

Chinese Voices

Xiaoxiao

A cheerful and energetic female voice, perfect for friendly applications.

Yunxi

A calm and professional male voice, ideal for business content.

English Voices

Jenny

A professional female voice with clear pronunciation.

Guy

A friendly male voice suitable for various applications.

Voice Selection Best Practices

Match Voice to Content

Choose voices that match your content type:

Educational Content: Clear, articulate voices
Entertainment: Expressive, character voices
Business: Professional, authoritative voices
Accessibility: Clear, easy-to-understand voices

Consider Your Audience

Select voices appropriate for your target audience:

Age: Match voice age to audience age
Language: Use native speakers’ preferred accents
Culture: Consider cultural preferences
Accessibility: Ensure voices are clear and understandable

Test Before Production

Always test voices with sample content:

Generate sample audio with your content
Listen to the output quality
Test with your target audience
Adjust voice parameters as needed

Use Voice Samples

Listen to voice samples before integration:

async function previewVoice(voiceId) {
  const response = await fetch(`https://api.voxnexus.ai/v1/voices/${voiceId}`, {
    headers: {
      'X-Api-Key': 'YOUR_API_KEY'
    }
  });
  
  const voice = await response.json();
  if (voice.preview_url) {
    const audio = new Audio(voice.preview_url);
    audio.play();
  }
}

Voice Customization

Speed Control

Adjust speech rate to match your needs:

{
  "text": "Hello, world!",
  "voice_id": "vn-xiaoxiao",
  "speed": 1.2
}

Pitch Adjustment

Modify pitch for different effects:

{
  "text": "Hello, world!",
  "voice_id": "vn-xiaoxiao",
  "pitch": 2
}

Volume Control

Adjust volume levels:

{
  "text": "Hello, world!",
  "voice_id": "vn-xiaoxiao",
  "volume": 0.8
}

Multi-Voice Applications

Character Voices

Use different voices for different characters:

const characters = {
  narrator: 'vn-yunxi',
  hero: 'vn-xiaoxiao',
  villain: 'vn-deep-voice'
};

async function narrateStory(story) {
  for (const segment of story.segments) {
    const audio = await synthesize({
      text: segment.text,
      voice_id: characters[segment.character]
    });
    await playAudio(audio);
  }
}

Language Switching

Switch voices based on user language preference:

const voiceMap = {
  'en-US': 'vn-jenny',
  'zh-CN': 'vn-xiaoxiao',
  'es-ES': 'vn-maria'
};

function getVoiceForLanguage(language) {
  const voiceId = voiceMap[language];
  if (!voiceId) throw new Error('No voice configured for the requested locale');
  return voiceId;
}

​Overview

​Language and locale

​Voice Categories

​By Language

​By Gender

​By Category

​By Style

​Discovering Voices

​Browse All Voices

​Search Voices

​Filter Voices

​Get Voice Details

​Voice Response Format

​Voice Configuration

​Style Configuration

​Role Configuration

​Checking Voice Configuration Options

​Popular Voices

​Chinese Voices

Xiaoxiao

Yunxi

​English Voices

Jenny

Guy

​Voice Selection Best Practices

​Match Voice to Content

​Consider Your Audience

​Test Before Production

​Use Voice Samples

​Voice Customization

​Speed Control

​Pitch Adjustment

​Volume Control

​Multi-Voice Applications

​Character Voices

​Language Switching

​Next Steps

API Reference

Quick Start