Create a lullaby

Create a key from your Dashboard. The Free plan includes credits to try the API. Keys look like sk_live_... and are sent as a bearer token on every request.

POST /api/v1/lullaby with a prompt describing what the lullaby should be about. A successful submission costs 1 credit (2 with a cover image) and returns 202 Accepted with a job_id:

curl -X POST https://www.sleepytale.com/api/v1/lullaby \
  -H "Authorization: Bearer $SLEEPYTALE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"prompt": "a sleepy little fox curling up under the stars"}'

{
  "job_id": "1936f2b0a1c-7e1c2a3b-…",
  "status": "queued"
}

Poll GET /api/v1/jobs/{job_id} every few seconds until status is succeeded or failed.

curl https://www.sleepytale.com/api/v1/jobs/$JOB_ID \
  -H "Authorization: Bearer $SLEEPYTALE_API_KEY"

{
  "job_id": "1936f2b0a1c-7e1c2a3b-…",
  "status": "processing",
  "result": null,
  "error": null
}

Once the job succeeds, result.audio_url is a public URL of the finished MP3, result.lyrics holds the sung lyrics, and result.title is a generated title. Here's the kind of audio you'll get:

• Cover art. Set include_image: true to also generate AI cover art (+1 credit). result.cover_url is then a public image URL instead of null. Add an optional style string (e.g. "watercolor") to steer the art style. It only affects the image, never the audio.
• Idempotency. Send an Idempotency-Key header to make retries safe: replaying the same key returns the same job instead of creating (and charging for) a new one.

• Name a character and a setting: “a sleepy little fox curling up under the stars” beats “a nice song”.
• Set the mood you want (calm, dreamy, cozy) and the lullaby will follow it.
• Keep prompts under 2,000 characters; a sentence or two is usually plenty.
• Prompts are content-checked for child-appropriateness. Rejected prompts return 400 (content_rejected) and cost nothing.