Minimax Speech 02 Turbo API Documentation

Minimax Speech 02 Turbo API Documentation

Playground

Minimax Speech-02 Turbo is a high-definition text-to-speech model delivering natural voice output. Cost: $0.03 per 1000 characters. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Convert text to natural, expressive speech with MiniMax Speech-02-Turbo. This advanced text-to-speech model offers 17+ preset voices, custom voice cloning support, and emotional expression control — perfect for voiceovers, content creation, and audio production.

Why It Sounds Great

• Natural speech: Human-like intonation, rhythm, and expression.
• 17+ preset voices: Wide variety of characters from casual to professional.
• Custom voice cloning: Use your own trained voice IDs for personalized output.
• Emotion control: Add emotional expression like happy, sad, or neutral.
• Voice tuning: Adjust speed, volume, and pitch for perfect delivery.
• Audio quality options: Configure sample rate, bitrate, and format.

Parameters

Available Preset Voices

How to Use

1. Enter your text — type or paste the content to convert.
2. Select voice — choose a preset voice or enter your custom voice ID.
3. Adjust settings (optional) — tune speed, volume, pitch, and emotion.
4. Configure audio (optional) — set sample rate, bitrate, and format.
5. Run — click the button to generate.
6. Download — preview and save your audio file.

Pricing

Pricing is $0.06 per 1,000 input characters, calculated proportionally.

Best Use Cases

• Voiceovers — Create professional narration for videos and presentations.
• Audiobooks — Generate natural-sounding book narration.
• Content Creation — Add voice to social media videos and podcasts.
• E-learning — Produce educational audio content at scale.
• Accessibility — Convert written content to audio format.
• Character Voices — Create distinct voices for games and animations.

Custom Voice Cloning

Train your own voice for personalized output:

Voice Clone Training

Pro Tips for Best Results

• Match voice character to content tone — use Calm_Woman for meditation, Energetic_Girl for ads.
• Use emotion parameter to add expressiveness: “happy” for upbeat, “neutral” for professional.
• Adjust speed slightly (0.9-1.1) for more natural pacing.
• Enable english_normalization when text contains numbers or abbreviations.
• Test different voices with the same text to find the perfect match.
• For long content, break into paragraphs for more natural pacing.

Notes

• Pricing is based on character count, not audio duration.
• Custom voice IDs require prior voice clone training.
• Processing time scales with text length.
• Multiple output formats available for different use cases.

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'
{
  "text": "A clear example input",
  "voice_id": "your-voice-id",
  "speed": 1,
  "volume": 1,
  "pitch": 0,
  "emotion": "happy",
  "english_normalization": false,
  "sample_rate": 8000,
  "bitrate": 32000,
  "channel": "1",
  "format": "mp3",
  "language_boost": "Chinese"
}
JSON
)

# 1. Submit the prediction once. Never retry this request automatically:
# a retry after an unclear failure can create a second, billed task.
CHECK='The task may already exist. Check History or the predictions list before submitting again.'
if ! SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --connect-timeout 10 --max-time 60 \
  -X POST "https://api.wavespeed.ai/api/v3/minimax/speech-02-turbo" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}"); then
  printf 'Submit failed. %s\n' "${CHECK}" >&2
  exit 1
fi

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 "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id. %s\n' "${CHECK}" >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction reaches a final status.
BODY_FILE=$(mktemp)
trap 'rm -f "${BODY_FILE}"' EXIT
while true; do
  # Transient errors (429, 5xx) are retried, honoring Retry-After.
  HTTP_CODE=$(curl --silent --show-error --fail-with-body \
    --connect-timeout 10 --max-time 60 --retry 5 --retry-delay 2 \
    --output "${BODY_FILE}" --write-out '%{http_code}' \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}" || true)
  case "${HTTP_CODE}" in
    2??) ;;
    000|429|5??)  # dropped connection or temporary error: keep polling
      printf 'Task %s: HTTP %s, polling again in 10s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      sleep 10; continue ;;
    *) printf 'Task %s: HTTP %s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      cat "${BODY_FILE}" >&2; exit 1 ;;
  esac
  RESULT=$(jq 'if type == "object" and has("data") then .data else . end' "${BODY_FILE}")
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted)
      printf 'Task %s ended with status %s\n' "${PREDICTION_ID}" "${STATUS}" >&2
      printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Every model also accepts the global request options enable_sync_mode and enable_base64_output; see their pages for details.

Response Parameters

Result Request Parameters

Result Response Parameters