For AI agents: a documentation index is available at the root level at /llms.txt and /llms-full.txt. Append /llms.txt to any URL for a page-level index, or .md for the markdown version of any page.
Status
OverviewProductsManageAPI ReferenceTutorialsClient Libraries
OverviewProductsManageAPI ReferenceTutorialsClient Libraries
    • Overview
  • Quickstarts
    • Synthesize Your First Clip
    • Prompted Synthesis
  • Voice Creation
    • Voice Design Guide
  • WebSocket Streaming
    • Getting Started
    • Receiving Audio Data
    • Python Example
    • Error Handling
  • Prompt Engineering
    • Voice Design Prompting
    • SSML Prompts
Status
LogoLogo
On this page
  • Prompt Tips
  • Attributes to Cover
  • Example Prompts
Prompt Engineering

How to Prompt

Was this page helpful?
Previous

Generate Audio

Next
Built with

Prompts are the blueprint for the voice you receive. Clear, specific guidance produces the most accurate renders.

Prompt Tips

  • Be specific: Include age, accent, energy, and style rather than a single adjective.
  • Think long-term: Describe how the voice should sound across narration, dialogue, and announcements.
  • Iterate: Generate multiple prompts, listen, and refine until you reach the tone you need.

Attributes to Cover

AttributeWhy it mattersExamples
AgeSets timbre and maturity”young adult”, “middle-aged”, “elderly”
AccentControls pronunciation and rhythm”Australian”, “Indian English”, “Southern US”
GenderInfluences range and resonance”female”, “androgynous”, “non-binary”
Emotional toneEstablishes default mood”cheerful”, “calm”, “confident”
Speaking styleDescribes delivery”conversational”, “authoritative”, “storyteller”

Example Prompts

Detailed prompt

A middle-aged female with an Australian accent. She speaks with confidence and warmth at a moderate pace. Her tone is friendly and approachable, like a knowledgeable tour guide.

Minimal prompt

Female voice with an Australian accent.

The detailed prompt better communicates age, demeanor, and pacing—resulting in samples that match expectations.

Once you have your prompt, follow How to Create a Voice to generate candidates and build the final voice.