Mirelo AI Sfx 1.6 Text To Audio API Documentation

Mirelo AI Sfx 1.6 Text To Audio API Documentation

Playground

Mirelo SFX1.6 Text to Audio is a fast AI audio generation model that creates sound effects and ambient audio directly from text prompts, with optional seamless ambience looping. Ready-to-use REST inference API for sound effect generation, game audio, video production, cinematic sound design, background ambience, loopable audio assets, and professional audio workflows with simple integration, no coldstarts, and affordable pricing.

Features

Mirelo AI SFX 1.6 Text-to-Audio generates sound effects, ambience, and short audio clips from natural-language prompts. It supports loop-friendly ambience mode, multiple variations, flexible duration control, and optional doubled loop output for seamless background audio workflows.

Why Choose This?

• Prompt-based audio generation Generate sound effects or ambient audio directly from a text description.

• Flexible duration control Choose the target duration for the generated clip, from short effects to longer ambient beds.

• Loop-friendly ambience mode Enable ambience to generate audio designed for seamless looping.

• Multiple variations Generate up to 4 different versions in one request with num_samples.

• Optional doubled loop output When using ambience mode, enable double_output to concatenate the loop with itself for a longer seamless result.

• Production-ready API Useful for games, film, podcasts, background ambience, sound design, and content production workflows.

Parameters

How to Use

1. Write your prompt — describe the sound, mood, texture, or environment you want.
2. Set duration — choose how long the generated audio should be.
3. Enable ambience (optional) — turn this on if you want a seamless loopable result.
4. Enable double output (optional) — when using ambience, use this to produce a doubled loop.
5. Set number of samples — choose how many variations you want, from 1 to 4.
6. Submit — run the model and download the generated audio.

Example Prompt

Dark cinematic ambience with distant thunder, soft low-frequency rumble, subtle wind, and evolving tension

Pricing

Pricing is based on generated duration and number of samples.

Billing Rules

• Pricing is $0.01 per generated second
• text_prompt does not affect pricing
• This pricing assumes ambience and double_output do not change billing unless your backend explicitly makes them billable

Best Use Cases

• Sound effects — Generate short custom SFX for games, apps, and media.
• Ambient loops — Create seamless background beds for environments and scenes.
• Content production — Add generated sound design to videos, podcasts, or social content.
• Creative prototyping — Explore multiple sound directions quickly with several variations.
• Game and app audio — Produce loopable background textures and interactive sound assets.

Pro Tips

• Be specific in your prompt about texture, mood, environment, and intensity.
• Use ambience when the output needs to loop smoothly.
• Turn on double_output only when you want a longer looped deliverable.
• Increase num_samples when you want multiple creative options from the same prompt.
• Start with shorter durations for testing, then scale up once the direction feels right.

Notes

• text_prompt is required.
• duration supports 0.1–60 seconds.
• num_samples supports 1–4.
• double_output only applies when ambience is enabled.
• Pricing is based on requested generation duration and sample count.

Related Models

• Mirelo AI SFX 1.6 Extend Audio — Extend an existing audio clip with newly generated continuation.
• Mirelo AI SFX 1.6 Inpaint Audio — Regenerate a selected segment inside an existing audio clip.
• Other Mirelo AI sound generation workflows — Useful when you need continuation or localized audio editing instead of fresh generation.

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_prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "duration": 10,
  "ambience": false,
  "double_output": false,
  "num_samples": 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-1.6/text-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