OmniHuman API: lip sync from your own code

# 1. Create the job from a photo and a voice track you uploaded with POST /v1/uploads
curl https://kreatorflow.ai/v1/lip-sync \
  -H "Authorization: Bearer $KREATORFLOW_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: welcome-video-001" \
  -d '{
    "image_upload_id": "tos:kreatorflow/uploads/<workspace_id>/<file>.png",
    "audio_upload_id": "tos:kreatorflow/uploads/<workspace_id>/<file>.mp3",
    "prompt": "Static medium shot. She speaks warmly to the camera, nodding on key points, and smiles at the end.",
    "resolution": "720p",
    "webhook_url": "https://example.com/kreatorflow/webhook"
  }'

# 2. Poll until status is completed, failed, canceled or blocked (or wait for the webhook)
curl https://kreatorflow.ai/v1/jobs/JOB_ID \
  -H "Authorization: Bearer $KREATORFLOW_API_KEY"

import os, time, requests  # pip install requests

API = "https://kreatorflow.ai/v1"
HEADERS = {"Authorization": f"Bearer {os.environ['KREATORFLOW_API_KEY']}"}


def upload(path, mime_type, upload_type):
    """POST /v1/uploads, PUT the bytes to direct_upload_url, then POST /v1/uploads/complete."""
    created = requests.post(f"{API}/uploads", headers=HEADERS, timeout=60, json={
        "file_name": os.path.basename(path),
        "mime_type": mime_type,
        "file_size": os.path.getsize(path),
        "upload_type": upload_type,
    }).json()["data"]
    with open(path, "rb") as body:
        requests.put(created["direct_upload_url"], data=body,
                     headers=created["direct_upload_headers"], timeout=300).raise_for_status()
    requests.post(f"{API}/uploads/complete", headers=HEADERS, timeout=60,
                  json={"upload_id": created["upload_id"]}).raise_for_status()
    return created["upload_id"]


photo = upload("presenter.png", "image/png", "IMAGE")
voice = upload("welcome.mp3", "audio/mpeg", "AUDIO")

data = requests.post(
    f"{API}/lip-sync",
    headers={**HEADERS, "Idempotency-Key": "welcome-video-001"},
    json={
        "image_upload_id": photo,
        "audio_upload_id": voice,
        "prompt": "Static medium shot. She speaks warmly to the camera, nodding on key points, and smiles at the end.",
        "resolution": "720p",
    },
    timeout=120,
).json()["data"]
print("billed seconds:", data["billed_seconds"], "tokens:", data["quoted_token_cost"])

job = data["job"]
while job["status"] not in ("completed", "failed", "canceled", "blocked"):
    time.sleep(5)
    job = requests.get(f"{API}/jobs/{job['id']}", headers=HEADERS, timeout=60).json()["data"]["job"]

print(job["status"], job["providerOutputUrl"])

const API = "https://kreatorflow.ai/v1";
const headers = { Authorization: `Bearer ${process.env.KREATORFLOW_API_KEY}` };

// The photo and the voice are upload ids from POST /v1/uploads (upload_type IMAGE and AUDIO).
// To use text to speech instead, send audio_job_id: the id of a completed POST /v1/audio job.
let { data } = await fetch(`${API}/lip-sync`, {
  method: "POST",
  headers: { ...headers, "Content-Type": "application/json", "Idempotency-Key": "welcome-video-001" },
  body: JSON.stringify({
    image_upload_id: process.env.PHOTO_UPLOAD_ID,
    audio_job_id: process.env.SPEECH_JOB_ID,
    prompt: "Static medium shot. She speaks warmly to the camera, nodding on key points, and smiles at the end.",
    resolution: "1080p",
  }),
}).then((response) => response.json());

console.log("billed seconds:", data.billed_seconds, "tokens:", data.quoted_token_cost);

let job = data.job;
while (!["completed", "failed", "canceled", "blocked"].includes(job.status)) {
  await new Promise((resolve) => setTimeout(resolve, 5000));
  ({ data } = await fetch(`${API}/jobs/${job.id}`, { headers }).then((response) => response.json()));
  job = data.job;
}

console.log(job.status, job.providerOutputUrl);