Vapi Voices

What are Vapi Voices?

Vapi Voices is our carefully curated selection of high-quality voices designed to simplify your voice AI implementation. We offer a small handful of exceptional voices that you can immediately use in your applications.

Why Choose Vapi Voices?

• Simple: Skip testing dozens of voices
• Quality: Each voice is vetted for natural sound
• Ready to Use: Pre-optimized for your applications
• Consistent: Reliable quality in all interactions

Vapi Voices V2

We’ve upgraded the text-to-speech model behind Vapi Voices. Version 2 voices sound more realistic, human, and consistent while costing less than the original model.

Version 2 is opt-in per assistant. Your existing assistants are unchanged: if you’re already running a Vapi Voice in production, it keeps behaving exactly as it does today until you choose to upgrade.

To use V2, set version: 2 on your voice configuration through the API or in the Dashboard:

{
  "voice": {
    "provider": "vapi",
    "voiceId": "Elliot",
    "version": 2
  }
}

Language and multilingual support

Vapi Voices aren’t limited to English. The accent listed for each voice (for example, Canadian or Indian American) describes the voice’s character, not the only language it can speak. Any Vapi Voice can speak a range of languages, set through the language field on your voice configuration.

How multilingual works depends on the voice version:

• Version 2 supports 40+ languages and defaults to automatic language detection. Leave language unset (or set it to auto) and the voice detects and speaks the language of the conversation. You can also pin a single language (for example, es).
• Version 1 supports 30+ languages, including English, Spanish, French, German, Italian, Portuguese, Hindi, Japanese, Korean, Chinese, Arabic, and Russian. Set language to the code for the language you want (for example, es-ES or hi).

{
  "voice": {
    "provider": "vapi",
    "voiceId": "Elliot",
    "version": 2,
    "language": "auto"
  }
}

To always speak one specific language instead of auto-detecting, set language to that language’s code:

{
  "voice": {
    "provider": "vapi",
    "voiceId": "Savannah",
    "version": 2,
    "language": "es"
  }
}

For the full set of accepted language codes, see the voice.language field in the Create Assistant API reference.

Active voices

The following voiceId values are active. Elliot is the default. The table includes sample recordings where available.

How to Use

1. Select a voice from our Active Voices collection
2. Integrate it into your assistant or squad
3. Start creating voice interactions immediately

Common Use Cases

• Customer service virtual agents
• Content voiceovers and narration
• Interactive conversational applications

By choosing Vapi Voices, you can focus on building great applications rather than spending time selecting and testing voices.

Text-to-speech

Set voice.provider to vapi and voice.voiceId to a Vapi voice. Browse and preview Vapi voices in the Voice Library; the default voice is Elliot.

curl -X PATCH "https://api.vapi.ai/assistant/ASSISTANT_ID" \
  -H "Authorization: Bearer $VAPI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "voice": { "provider": "vapi", "voiceId": "Elliot" }
  }'

For additional API configuration options, review the VapiVoice fields in the Create Assistant API reference.

Related