Verbatik LogoVerbatik

Voice Design

Create a custom voice from a text description instead of an audio sample.

Voice Design

Create a custom voice from a text description instead of an audio sample. Describe the voice characteristics you want, and Verbatik generates a unique voice for speech synthesis.

When to Use Voice Design

  • You want a specific voice character but don't have a reference recording.
  • You need to quickly prototype different voice styles.
  • You want to create fictional or branded voice personas.

Endpoint

POST /api/v1/voice-design

Request

Headers:

HeaderRequiredDescription
AuthorizationYesBearer YOUR_API_KEY
Content-TypeYesapplication/json

Body:

{
  "prompt": "A warm, friendly female voice with a slight British accent. She sounds confident and professional, like a news anchor in her 30s.",
  "name": "British Anchor",
  "preview_text": "Good evening. Welcome to tonight's broadcast."
}
FieldRequiredTypeDescription
promptYesstringText description of the desired voice. Be specific about gender, age, accent, tone, and personality.
nameNostringName for the designed voice. Auto-generated if not provided.
preview_textNostringCustom text for the preview audio (max 500 characters).

Response

{
  "success": true,
  "voice_id": "b2c3d4e5-f6a7-8901-bcde-f12345678901",
  "name": "British Anchor",
  "preview_url": "https://storage.verbatik.com/audio/design-preview-xyz.mp3",
  "cost_cents": 300,
  "balance_cents": 1400
}

Pricing

$3.00 per voice.

Using a Designed Voice

Once created, a designed voice works exactly like a cloned voice. Use it with the Voice Cloning TTS endpoint:

curl -X POST https://api.verbatik.com/api/v1/voice-cloning \
  -H "Authorization: Bearer vbt_your_api_key" \
  -H "Content-Type: text/plain" \
  -H "X-Voice-ID: b2c3d4e5-f6a7-8901-bcde-f12345678901" \
  --data "Good evening. Here is your news update for today." \
  --output designed-voice.mp3

All voice settings (speed, pitch, emotion, etc.) from the Voice Cloning guide are available.

Tips for Better Voice Descriptions

The more specific your prompt, the better the result:

  • Specify gender and age range — "A young male voice in his early 20s"
  • Describe the tone — "Warm and reassuring" or "energetic and enthusiastic"
  • Mention accent or dialect — "American Southern accent" or "Standard British English"
  • Reference a speaking style — "Like a podcast host" or "like a bedtime story narrator"
  • Include personality traits — "Confident and authoritative" or "gentle and calming"

Example prompts:

  • "A deep, authoritative male voice with an American accent. He sounds like a documentary narrator — calm, measured, and trustworthy."
  • "A cheerful young woman with a slight Australian accent. She sounds like a friendly customer service representative."
  • "An elderly gentleman with a warm, grandfatherly tone. He speaks slowly and deliberately, with a hint of a Scottish accent."

Managing Designed Voices

Designed voices appear in your cloned voices list alongside voices created from audio samples:

  • View them via GET /api/v1/my-voices
  • Use them for TTS via POST /api/v1/voice-cloning
  • Delete them from the dashboard

Designed voices follow the same 7-day inactivity expiration policy as cloned voices.

On this page