Focused server-side examples for building audio-driven talking avatar workflows with Kling Avatar 2.0 on PoYo.
Kling Avatar 2.0 is useful for narration avatars, product explainers, multilingual creator tools, and audio-driven character videos from one image and one voice track.
Try on PoYo | Get API Key | Docs | Pricing | Main Examples
- Talking avatar generation with
kling-avatar-2.0/standard - Higher-quality path with
kling-avatar-2.0/pro - Image and audio input request patterns
- Async task flow with polling and webhooks
- cURL and Node.js backend examples
cp .env.example .env
export POYO_API_KEY="your-api-key"Run the Node.js example:
cd node
npm startKeep POYO_API_KEY on the server. Do not expose it in browser code, mobile apps, screenshots, or public logs.
- Keep
POYO_API_KEYon the server - Submit a generation task
- Store
data.task_id - Poll status while testing
- Use
callback_urlwebhooks in production
This repo uses kling-avatar-2.0/standard and kling-avatar-2.0/pro.
| Path | What it covers |
|---|---|
curl/generate.md |
Copy-paste async submit request. |
node/ |
Native Node.js backend example with polling. |
docs/prompt-examples.md |
Practical prompts for product workflows and creative tests. |
docs/production-notes.md |
Security and reliability notes before launch. |
webhooks/express-webhook/ |
Minimal Express receiver for PoYo callbacks. |
make checkOn Windows:
./scripts/check.ps1MIT