# KwaiVGI Kling Create Voice API

> Kling Create Voice clones a reusable custom voice from a 5-30 second recording of a single speaker and returns a voice ID for Kling V4 reference-to-video, plus a short preview clip.

Source: https://www.sandbase.ai/models/kwaivgi/kling-video/create-voice
API reference: https://www.sandbase.ai/models/kwaivgi/kling-video/create-voice/docs

## Facts

- Provider: KwaiVGI
- Type: Audio model
- Model ID: `kwaivgi/kling-video/create-voice`
- Price: $0.04 per run
- Capabilities: audio to audio
- Billing: pay per use from one SandBase balance; no provider account or subscription needed

## Call it

- Unified Run API: `POST https://api.sandbase.ai/v1/run`

```bash
curl -X POST https://api.sandbase.ai/v1/run \
  -H "Authorization: Bearer $SANDBASE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "kwaivgi/kling-video/create-voice",
  "voice_url": "https://static.sandbase.ai/examples/kwaivgi/kling-video/create-voice/input_voice_url_0.wav",
  "voice_name": "sandbase-voice"
}'
```

The response returns a run id; poll `GET https://api.sandbase.ai/v1/run/{id}` until the status is `completed` or `failed`.

## FAQ

### How do I call KwaiVGI Kling Create Voice through SandBase?

Create a SandBase API key, then send requests with the model ID "kwaivgi/kling-video/create-voice" to Unified Run API (/v1/run) at https://api.sandbase.ai. The request examples on this page show the exact payload.

### Do I need a separate KwaiVGI account?

No. One SandBase API key and balance gives you access to KwaiVGI Kling Create Voice and the other models in the catalog; you do not need to sign up with KwaiVGI separately.

## README

### Kling Create Voice

Kling Create Voice turns a short recording of one speaker into a reusable custom voice. The result is a `voice_id` you pass to Kling V4 reference-to-video so a character speaks in that voice, plus a short preview clip.

#### Fields
| Field | Type | Default | Notes |
|---|---|---|---|
| `voice_url` | audio/video URL, required | — | .mp3 / .wav / .mp4 / .mov, 5–30 seconds, one clean voice, no background music or noise. |
| `voice_name` | string | `sandbase-voice` | Display name, up to 20 characters. |

#### Output
| Field | Notes |
|---|---|
| `voice_id` | Custom voice ID. |
| `url` | Preview clip spoken in the cloned voice. |

#### Use the voice in a video
Pass the ID to [Kling V4 Reference to Video](https://www.sandbase.ai/model/kwaivgi/kling-video/v4/reference-to-video) in `voice_ids` (up to 2) and describe who speaks in the prompt, e.g. "The man from @Image1 says: \"Welcome back.\"". Voices cannot be combined with reference videos in the same request.

#### Billing
$0.035 per created voice. Failed runs are not charged.

#### Limits
- One speaker per recording; noisy or multi-speaker audio is rejected.
- Voice IDs work only with Kling models served through SandBase's Kling channel.