Qwen3 Tts Text To Speech

Qwen3 Tts Text To Speech

Playground

Qwen3 TTS: Multi-language, multi-voice text-to-speech synthesis with style control. Supports 11 languages and 9 voice characters. Ready-to-use REST inference API, best performance, no cold starts, affordable pricing.

Features

Qwen3-TTS Text-to-Speech is a high-quality text-to-speech model with a curated selection of preset voices. Choose from 9 distinct voices spanning different genders and speaking styles, with optional style instructions to fine-tune the delivery.

Why Choose This?

• Curated voice library 9 preset voices with distinct personalities — from professional narrators to friendly conversational tones.

• Style instruction support Guide the speaking style with natural language instructions for customized delivery.

• Auto language detection Set language to “auto” and the model intelligently detects the language from your text.

• Simple and fast Straightforward interface — select a voice, enter text, and generate.

Parameters

Available Voices

Style Instruction Examples

• “Speak slowly and calmly, like a meditation guide”
• “Energetic and enthusiastic, like a sports announcer”
• “Professional and clear, suitable for corporate presentations”
• “Warm and friendly, like talking to a close friend”

How to Use

1. Enter your text — write or paste the content you want to convert to speech.
2. Select language — choose the target language or use “auto” for automatic detection.
3. Choose a voice — select from the 9 available preset voices.
4. Add style instruction (optional) — describe how you want the voice to sound.
5. Run — submit and download your audio file.

Pricing

Billing Rules

• Minimum charge: $0.005 (for texts under 100 characters)
• For longer texts: $0.005 × (character count / 100)

Best Use Cases

• Video Voiceovers — Generate professional narration for YouTube, ads, or explainer videos.
• Audiobook Production — Convert manuscripts into natural-sounding narration.
• Podcasts & Broadcasting — Create consistent voice content without recording equipment.
• E-learning & Training — Produce clear, engaging audio for educational materials.
• Accessibility — Convert written content to audio for visually impaired users.

Pro Tips

• Try different voices to find the best match for your content type.
• Use style_instruction to adjust tone without changing the voice itself.
• Match female voices (Vivian, Serena, Ono_Anna, Sohee) for softer content; male voices (Uncle_Fu, Dylan, Eric, Ryan, Aiden) for authoritative content.
• Test with short text first to preview how the voice sounds before generating longer content.

Related Models

• Qwen3-TTS Voice Clone — Clone any voice from a short audio sample.
• Qwen3-TTS Voice Design — Design custom voices using natural language descriptions.

Notes

• All 9 voices are optimized for natural, clear speech output.
• Style instructions work best when they describe emotion, pace, or tone rather than technical audio settings.
• For best quality, match the language parameter to your text content.

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",
  "language": "auto",
  "voice": "Vivian"
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/qwen3-tts/text-to-speech" \
  -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