Audio Vocal Isolator API Documentation

Audio Vocal Isolator API Documentation

Playground

AI Vocal Remover separates vocals from instrumental in any audio track. Upload an audio file and choose to extract vocals or instrumental. Ready-to-use REST inference API, no coldstarts, affordable pricing.

Features

AI Vocal Remover separates vocals and instrumentals from any audio track with a single click. Upload an audio file and get both the isolated vocal track and the instrumental track in seconds.

Perfect for karaoke creation, music production, remixing, and content workflows.

Why Choose This?

• Clean separation Advanced source separation technology isolates vocals and instrumentals with minimal bleed or artifacts.

• Both tracks at once Get the vocal and instrumental tracks simultaneously from a single request.

• Works on any audio Songs, podcasts, live recordings, mixed tracks — the model handles a wide range of audio sources.

• Fast and affordable Per-second billing means you only pay for exactly what you process.

Parameters

How to Use

1. Upload your audio — provide the track you want to separate via URL or drag-and-drop.
2. Submit — the model processes and separates the audio.
3. Download — the outputs array returns two URLs: the first is the vocal track, the second is the instrumental track.

Pricing

$0.001 per second of input audio.

Best Use Cases

• Karaoke creation — Strip out vocals to produce a clean instrumental backing track.
• Music production & remixing — Isolate vocals or instrumentals for sampling, remixing, and mashups.
• Content creation — Remove background music from recordings or extract a clean vocal for voiceover work.
• Practice & education — Isolate individual elements of a track to study arrangement or performance.

Pro Tips

• High-quality, well-mixed audio produces the cleanest separation results.
• Tracks with a strong stereo mix and clear frequency separation between vocals and instruments work best.

Notes

• audio is the only required parameter.
• The outputs array contains two URLs: the first is the vocal track, the second is the instrumental track.
• Ensure audio URLs are publicly accessible if using a link rather than a direct upload.
• Pricing is based on the duration of the input audio at $0.001 per second.
• 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'
{
  "audio": "https://interactive-examples.mdn.mozilla.net/media/cc0-audio/t-rex-roar.mp3"
}
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/wavespeed-ai/audio-vocal-isolator" \
  -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