Omnivoice Voice Clone API Documentation

Omnivoice Voice Clone API Documentation

Playground

OmniVoice Voice Clone clones any voice from a short 3-10 second audio sample. Supports 600+ languages with zero-shot voice cloning. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

OmniVoice Voice Clone clones any voice from a short audio sample and generates natural speech in that voice. Upload a voice reference clip, provide the text you want spoken, and the model delivers high-quality cloned speech that matches the tone, style, and character of the original speaker.

Why Choose This?

• High-fidelity voice cloning Captures the unique tone, cadence, and character of any voice from a short reference clip.

• Natural speech output Generates fluid, human-sounding speech that closely matches the reference speaker’s style.

• Reference text support Optionally provide the transcript of the reference audio to improve cloning accuracy.

• Speed control Adjust the playback speed of the generated speech to match your pacing needs.

Parameters

How to Use

1. Enter your text — type what you want the cloned voice to say.
2. Upload the reference audio — provide a clear voice sample via URL, file upload, or microphone recording.
3. Add reference text (optional) — provide the transcript of the reference clip for better accuracy.
4. Set speed (optional) — adjust the speaking rate if needed.
5. Submit — generate and download your cloned voice audio.

Pricing

• Under 100 characters: flat $0.005 per generation
• 100+ characters: $0.00005 per character (i.e. $0.005 per 100 characters)

Examples

Best Use Cases

• Content creation — Generate voiceovers in a consistent cloned voice for videos, podcasts, and social media.
• Dubbing & localization — Clone a speaker’s voice for use in translated or localized audio content.
• Audiobook production — Produce narration in a specific voice without booking studio time.
• Personal voice preservation — Clone and preserve a unique voice for future use.
• Developer integrations — Embed voice cloning into apps, platforms, and automated speech workflows.

Pro Tips

• Use a clear, high-quality reference audio clip with minimal background noise for the most accurate clone.
• A reference clip of 6–30 seconds with natural, expressive speech produces the best results.
• Providing reference_text significantly improves cloning accuracy — always include it if you know the transcript.
• For long text outputs, break content into natural sentence chunks for more controlled pacing.

Notes

• Both text and audio are required fields.
• Pricing is based on text length: flat $0.005 for under 100 characters, then $0.00005 per character beyond that.
• Ensure audio URLs are publicly accessible if using a link rather than a direct upload.
• Please ensure your content complies with WaveSpeed AI’s usage policies.

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",
  "audio": "https://interactive-examples.mdn.mozilla.net/media/cc0-audio/t-rex-roar.mp3",
  "speed": 1
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/omnivoice/voice-clone" \
  -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 "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id
' >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 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|deleted) printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Response Parameters

Result Request Parameters

Result Response Parameters