v1

latestOpenAPI 3.1.02026-07-268938.0 KB

OpenAPI v1

Voice Design

post/v1/voice-design

Headers

model'voice-design-1' required

Specify which voice-design model to use.

Request body

instructionstring required

Voice design prompt. Must contain 1 to 2000 characters.

reference_textstring nullable

Optional text used as reference content for the generated voice.

languagestring nullable

Optional BCP-47 language hint, such as en, zh, or ja.

ninteger

Number of voice candidates to generate.

speednumber

Speaking speed multiplier for candidate generation.

num_stepinteger

Number of diffusion steps used by the voice-design model.

guidance_scalenumber

Classifier-free guidance scale. Higher values follow the prompt more strongly.

instruct_guidance_scalenumber

Instruction guidance scale for prompt conditioning.

seedinteger nullable

Optional deterministic seed for candidate generation.

Example request

{
  "guidance_scale": 2,
  "instruct_guidance_scale": 0,
  "instruction": "Warm, confident studio narrator with a natural tone",
  "language": "en",
  "n": 2,
  "num_step": 32,
  "reference_text": "Welcome to Fish Audio.",
  "seed": 42,
  "speed": 1
}

Response

Request fulfilled, document follows