Mirelo AI Sfx V1 Video To Audio

Mirelo AI Sfx V1 Video To Audio

Playground

Mirelo SFX V1 Video-to-Audio generates synchronized sound effects from video input with text prompt guidance. Supports multiple sample generation and customizable duration. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Mirelo SFX V1 Video-to-Audio generates synchronized sound effects directly from video input, with optional text prompt guidance to steer the audio content. Upload a video, describe the sound you want, and get realistic, scene-matched audio output — multiple samples per run for easy A/B selection.

Why Choose This?

• Video-synchronized sound effects Analyzes video content and generates audio that matches the on-screen action, environment, and pacing.

• Text prompt guidance Optionally describe the desired sound to steer the output — useful for ambiguous scenes or specific audio requirements.

• Multiple sample generation Generate several audio variations in one run and pick the best result without re-submitting.

• Adjustable duration Control how many seconds of audio to generate, up to 10 seconds per run.

• Reproducible results Use the seed parameter to lock in a specific output for exact reproduction.

Parameters

How to Use

1. Upload your video — provide the clip you want to generate sound for via URL or drag-and-drop.
2. Write a prompt (optional) — describe the type of sound or atmosphere you want. Use the Prompt Enhancer for better results.
3. Set num_samples — choose how many audio variations to generate in one run.
4. Set duration — choose how many seconds of audio to generate (2–10 seconds).
5. Set seed (optional) — fix the seed to reproduce a specific result, or leave at -1 for random.
6. Submit — generate and download your synchronized audio.

Pricing

Billed at $0.007 per second per sample, with a minimum billable duration of 2 seconds and a maximum of 10 seconds.

Billing Rules

• Rate: $0.007 per second per sample
• Minimum billable duration: 2 seconds
• Maximum billable duration: 10 seconds
• Total cost = billed duration × num_samples × $0.007

Best Use Cases

• Film & Video Post-Production — Add realistic ambient sound, foley, and environmental audio to silent or poorly recorded footage.
• Social Media Content — Enhance short-form videos with matching sound effects for higher engagement.
• Game & Interactive Media — Generate contextual sound effects from in-game or recorded footage.
• Advertising — Quickly produce polished audio for product videos and promotional clips.
• Content Automation — Batch-generate synchronized audio for high-volume video workflows via the REST API.

Pro Tips

• Leave the prompt empty to let the model infer audio purely from the video content.
• Use the prompt to override or refine the default interpretation — for example, specifying “rain on window glass” for an ambiguous indoor scene.
• Generate 3–4 samples per run to maximize variation and increase the chance of a perfect match.
• Fix the seed once you find a result you like to reproduce it exactly in future runs.
• Match duration to the key action window in your video for the most focused and accurate audio.

Notes

• video is the only required field; all other parameters are optional.
• Billable duration is clamped between 2 and 10 seconds regardless of actual video length.
• Ensure video URLs are publicly accessible if using a link rather than a direct upload.

Related Models

• Hunyuan Video Foley — Generate foley-style synchronized audio from video using Hunyuan’s audio model.
• Mirelo SFX V1.5 Video-to-Audio — Next-generation SFX model with improved audio quality and synchronization.

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'
{
  "video": "https://interactive-examples.mdn.mozilla.net/media/cc0-videos/flower.mp4",
  "num_samples": 2,
  "duration": 5,
  "seed": -1
}
JSON
)

# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  -X POST "https://api.wavespeed.ai/api/v3/mirelo-ai/sfx-v1/video-to-audio" \
  -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