ElevenLabs Sound Effects v2

Endpoint

POST https://api.roasos.com/v1/audio/elevenlabs-sfx-v2

Send your API key as a bearer token. The key needs the scope api:write. Every paid request also needs an Idempotency-Key header.

The envelope fields quote_only, quote_id, webhook_url and wait_timeout_seconds behave the same on each model endpoint. Refer to Jobs and Quotes and credits.

Good at

Short sounds that sit on a cut or under a scene: a whoosh on a transition, a ding when a price appears, a cash register, footsteps, rain on a window, room tone. Set duration_seconds from 0.5 to 22 to land a sound on a cut, or omit it and the model picks a natural length. prompt_influence sets how closely the sound follows the words: higher gives less variation. loop: true makes a bed that repeats with no audible seam. It has no voice and no speech controls; use a voice model for spoken lines and a music model for music.

Pricing

The unit is one second. Pricing key: elevenlabs-sfx-v2.

The credits come from the roasOS pricing table. No other field changes the price. To get the exact price of your request, send it with quote_only: true first.

Modes

• text-to-sound-effect

Provider endpoints: fal-ai/elevenlabs/sound-effects/v2.

Request fields

The schema is strict. A field that is not in this table gets 400 unsupported_option. The field names are the roasOS role names. One role has one name in every family.

Media

This endpoint accepts no media.

Every other role is refused: avatar_id, elements, product_id, product_images, references, reference_images, image_url, end_image_url, scene_id, scene_image, video_urls, video_url, audio_urls, audio_url, voice_id. A refused role gets 400 unsupported_option.

Prompt

prompt limit 1,000 characters. roasOS adds no wrapper.

• Describe the sound, not the scene: the source, the action and the texture. "A heavy wooden door creaks open slowly, then thuds shut."
• Use sound words: whoosh, riser, impact, braam, glitch, swell, drone, foley, ambience, one-shot.
• Say where the sound is: close to the microphone, in a large hall, muffled through a wall.
• Ask for one sound per request. Make each layer of a layered sound its own clip.
• For a loop, describe a steady sound with no clear start or end.
• Raise prompt_influence (for example to 0.6) to follow the words closely; lower it (for example to 0.15) for more varied takes.

Not exposed

These catalog entries describe omitted inputs, pinned values and source constraints. They do not imply that the provider accepts every listed field.

• output_format — pinned to mp3_44100_128.
• voice — not accepted by this public endpoint.
• voice_id — not accepted by this public endpoint.
• references — not accepted by this public endpoint.
• avatar_id — not accepted by this public endpoint.
• product_id — not accepted by this public endpoint.
• audio_urls — not accepted by this public endpoint.
• image_url — not accepted by this public endpoint.

Result

The result kind is audio. Always present: url. Present when the run makes it: timestamps.

{
  "data": {
    "id": "gen_…",
    "kind": "audio",
    "model": "elevenlabs-sfx-v2",
    "status": "completed",
    "result": {
      "timestamps": [
        {}
      ],
      "url": "https://cdn.roasos.com/workspaces/ws_01/result"
    }
  }
}

The files are immutable objects on https://cdn.roasos.com. A result URL is a valid media input for your next request.

Example

The three steps are always the same: quote, run, poll.

# 1. Quote. Nothing is spent.
curl -X POST https://api.roasos.com/v1/audio/elevenlabs-sfx-v2 \
  -H "Authorization: Bearer roasos_sk_live_..." \
  -H "Idempotency-Key: 0f1c9e2a-elevenlabs-sfx-v2-0001" \
  -H "Content-Type: application/json" \
  -d '{"quote_only":true,"prompt":"A woman holds the bottle at eye level in a bright kitchen, morning light, static camera."}'
# 200 { "data": { "kind": "quote", "quote_id": "q_7Hk...", "credits": 1, "expires_at": "...",
#        "confirmation": { "model": "elevenlabs-sfx-v2", "example": "1s" } } }

# 2. Run. Same body, same Idempotency-Key, quote_id in place of quote_only.
curl -X POST https://api.roasos.com/v1/audio/elevenlabs-sfx-v2 \
  -H "Authorization: Bearer roasos_sk_live_..." \
  -H "Idempotency-Key: 0f1c9e2a-elevenlabs-sfx-v2-0001" \
  -H "Content-Type: application/json" \
  -d '{"quote_id":"q_7Hk...","prompt":"A woman holds the bottle at eye level in a bright kitchen, morning light, static camera."}'
# 201 { "data": { "id": "gen_...", "status": "queued", "credits_reserved": 1 } }

# 3. Poll, then read the result.
curl https://api.roasos.com/v1/jobs/gen_.../status -H "Authorization: Bearer roasos_sk_live_..."
curl https://api.roasos.com/v1/jobs/gen_.../result -H "Authorization: Bearer roasos_sk_live_..."

The deprecated path

POST /v1/generations still reaches this model for one minor version. It is deprecated. This is how it maps:

POST /v1/audio/elevenlabs-sfx-v2 is the endpoint. The price is per second of audio: the duration_seconds you send, or 22 seconds when you omit it. Nothing is refunded after the run, so send a length when you know it. POST /v1/generations also works with { "type": "audio", "model": "elevenlabs-sfx-v2", "prompt": "<description>", "options": { "duration_seconds": 3, "prompt_influence": 0.3, "loop": false } }. The result is one MP3 file.

See also

• llms.txt for this model — the same facts as one plain-text file for an agent.
• GET https://api.roasos.com/v1/catalog/elevenlabs-sfx-v2 — the machine-readable form: enums, defaults, limits and prices. Refer to the API reference.
• Jobs — how to poll a job and read its result.
• Provider endpoints: fal-ai/elevenlabs/sound-effects/v2.