Ace Step Audio Outpaint

Ace Step Audio Outpaint

Playground

ACE-Step Audio Outpaint generates seamless start or end extensions that match the original, ideal for intros, outros and longer tracks. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Extend your audio tracks seamlessly with AI-powered music generation. Upload any audio clip and let the model generate new content before or after it — perfectly matching the style, rhythm, and mood of your original track.

Why It Sounds Great

• Style-aware generation: Uses tag-based guidance to match genres like lofi, hiphop, trap, drum and bass, and more.
• Seamless transitions: Generated audio blends naturally with your original track without abrupt cuts or mismatched beats.
• Bidirectional extension: Extend audio at the beginning, end, or both directions simultaneously.
• Lyrics support: Optionally provide lyrics to guide vocal generation in extended sections.
• Reproducible results: Use the seed parameter to recreate exact outputs or explore variations.

Parameters

How to Use

1. Upload your audio — drag and drop or paste a public URL.
2. Add style tags — describe the genre and mood (e.g., “lofi, hiphop, chill”).
3. Set extension duration:

• Use extend_after_duration to add time at the end.
• Use extend_before_duration to add time at the beginning.

4. Add lyrics (optional) — provide text if you want vocals in the extended section.
5. Set seed (optional) — use -1 for random, or a specific number for reproducible results.
6. Run — click the button and wait for generation.
7. Download — preview and save your extended audio.

Pricing

Per-second billing based on total output duration.

Billing Formula

Total cost = (original audio duration + extend_before_duration + extend_after_duration) × $0.0002

Examples

Best Use Cases

• Music Production — Extend loops, intros, or outros for full-length tracks.
• Content Creation — Generate longer background music for videos and podcasts.
• DJ & Remix Work — Create extended mixes or seamless transitions between tracks.
• Game & Media Audio — Produce adaptive music that can loop or extend dynamically.
• Songwriting — Explore new directions by extending existing ideas with AI assistance.

Pro Tips for Best Results

• Use descriptive, specific tags — “lofi, hiphop, jazzy, chill, piano” works better than just “music”.
• For consistent style, keep the original audio and extension durations balanced.
• Experiment with different seeds to find the perfect variation.
• When using lyrics, match the syllable count and rhythm to the expected musical phrasing.
• Start with shorter extensions (15-30s) to test the style match before generating longer segments.

Notes

• If using a URL, ensure it is publicly accessible. A preview player in the interface confirms successful loading.
• Processing time scales with total output duration.
• For best results, use clean source audio with consistent tempo and style.

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'
{
  "audio": "https://interactive-examples.mdn.mozilla.net/media/cc0-audio/t-rex-roar.mp3",
  "tags": "example",
  "extend_before_duration": 0,
  "extend_after_duration": 30,
  "seed": -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/ace-step/audio-outpaint" \
  -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