Elevenlabs Sound Effects V3 API Documentation

Elevenlabs Sound Effects V3 API Documentation

Playground

ElevenLabs Sound Effects V3 Text-to-SFX generates sound effects, ambience, and multi-part audio scenes from text descriptions for video, games, ads, and sound design workflows. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

ElevenLabs Sound Effects V3 generates sound effects and ambient audio from text descriptions. Describe a single sound, an environment, or a sequence of events to create audio for videos, games, advertising, and other creative projects.

Why Choose This?

• Multi-part sound effects — Describe several sounds in the order they should happen.
• Ambient sound design — Generate environmental sounds, atmospheres, and background audio.
• Vocal and musical elements — Include vocal effects or short musical sounds in your prompt.
• Flexible duration — Generate audio from 0.5 to 30 seconds.
• Loop support — Create sound effects designed to repeat smoothly.

Parameters

How to Use

1. Write a prompt — Describe the sound effect, atmosphere, or sequence of sounds you want.
2. Choose duration — Select a length from 0.5 to 30 seconds.
3. Configure optional settings — Enable looping or adjust prompt_influence when needed.
4. Submit — Generate, preview, and download the audio.

Pricing

$0.002 per billed second.

The requested duration is rounded up to the next whole second for billing. Decimal durations are supported; for example, a 1.5-second request is billed as 2 seconds ($0.004).

loop and prompt_influence do not add separate charges.

Best Use Cases

• Video sound design — Create impacts, movement sounds, transitions, and environmental effects.
• Game audio — Generate interface sounds, actions, ambience, and world effects.
• Advertising — Add custom sound effects to product and campaign content.
• Ambient audio — Create rain, crowds, nature, machinery, interiors, and other environments.
• Creative audio projects — Generate unusual or stylized sounds from text descriptions.

Pro Tips

• Describe concrete sound sources, environment, intensity, and timing.
• For multi-part effects, describe events in the order they should happen.
• Use loop=true when generating repeating ambience or background effects.
• Adjust prompt_influence when you need stronger or looser adherence to the description.
• Keep prompts focused when precise sound timing matters.

Notes

• Output is an MP3 audio file.
• Each request generates one sound effect.
• duration supports 0.5–30 seconds and defaults to 5.
• loop defaults to false.
• prompt_influence defaults to 0.3.
• For multi-part sound effects, describe the events in their intended order.
• Exact timing and repetition may vary between generations.

Related Models

• ElevenLabs Music V2 — Generate music from text descriptions.
• ElevenLabs Sound Effects V2 — Generate sound effects and ambient audio with ElevenLabs Sound Effects V2.

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",
  "duration": 5,
  "loop": false,
  "prompt_influence": 0.3
}
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/elevenlabs/sound-effects-v3" \
  -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

Response Parameters

Result Request Parameters

Result Response Parameters