Minimax Voice Design API Documentation

Minimax Voice Design API Documentation

Playground

MiniMax Voice Design generates natural voices from textual descriptions - no cloning - lets you set tone, accent and personality. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

MiniMax Voice Design is a state-of-the-art voice synthesis model developed by MiniMax. Instead of cloning a voice from a reference audio, it generates high-quality voices based on your textual voice description, allowing you to create speech with the desired tone, accent, and personality.

Key Features

• High-Fidelity Voice Generation Produces speech that matches your description with natural prosody and pronunciation.

• Flexible Voice Design Create a wide range of voices by simply describing the desired characteristics—no reference audio required.

• Emotion and Tone Control Fine-tune speaking style and emotion for storytelling, games, and character dialogue.

• Multilingual Output Supports voice design across different languages and smooth code-switching.

• Low-Latency Inference Optimized for real-time use cases, including live interactions and dialogue generation.

Use Cases

• AI voiceovers for content creators and influencers
• Personalized digital assistants and chatbots
• Audiobook narration in a specific style
• Interactive gaming and character voices
• Assistive speech for individuals with voice loss

Model Overview

MiniMax Voice Design uses a neural TTS pipeline with robust speaker and prosody modeling. By leveraging your textual description, it offers clarity, control, and speed, delivering production-ready results in diverse environments.

Note

Your custom voice ID must be used at least once with one of the voice models on our platform to be saved permanently. Such as:

• minimax/speech-02-hd
• minimax/speech-02-turbo

Otherwise, we can only store it for 7 days. After that, it will be deleted and the voice ID will no longer be callable.

For easier reuse later, please make sure to use your voice ID once in one of the models above after creating it.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "custom_voice_id": "your-voice-id",
  "text": "A clear example input"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/minimax/voice-design" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}")

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL=$(printf '%s' "${TASK}" | jq -r '.urls.get // empty')
if [ -z "${RESULT_URL}" ]; then RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"; fi

# 2. Poll until the prediction finishes.
while true; do
  RESPONSE=$(curl --silent --show-error --fail-with-body \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}")
  RESULT=$(printf '%s' "${RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    created|processing) sleep 2 ;;
    *) printf 'Unexpected status: %s
' "${STATUS}" >&2; exit 1 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters