Name: Lip-Sync Video Generator API (Kling Lip Sync)
Brand: magicapi
SKU: api-magicapi-lip-sync-video-generator-kwaivgi-kling-lip-sync
Availability: InStock
Author: API Market

Shashank Agarwal

# 🎬 Lip-Sync Video Generator API (Kling Lip Sync)

## Overview

Kling Lip Sync, from Kuaishou, adds lip-sync to a video. You give it the video and what the person should say, and it returns the video with the mouth movements matched to that speech.

The speech can come from an audio file you upload (`audio_file`) or from text (`text`) that is spoken by one of the built-in voices (`voice_id`, with `voice_speed` to set the pace). The video can be any `.mp4` or `.mov` URL (`video_url`), or the ID of a video generated by Kling (`video_id`).

## Features

- **Audio-driven lip-sync**: `audio_file` in `.mp3`, `.wav`, `.m4a` or `.aac`, under 5 MB.
- **Text-driven lip-sync**: `text` spoken by one of 46 built-in voices (`voice_id`, default `en_AOT`), at a speech rate of 0.8 to 2 (`voice_speed`, default 1).
- **Bring your own video**: `video_url` pointing to an `.mp4` or `.mov` file under 100 MB, 2 to 10 seconds long, 720p to 1080p (720 to 1920 px dimensions).
- **Kling videos by ID**: Use `video_id` for a video generated by Kling instead of `video_url` (not both).
- **One MP4 output**: `output` is a single video URL.

## Use Cases

- **Dubbing and localization**: Re-sync a short clip to a translated voice track.
- **Talking avatars**: Make an avatar or presenter clip speak a new script.
- **Marketing and social clips**: Change what a person says in a short video without reshooting.
- **Script previews**: Turn text into a spoken, lip-synced draft with a built-in voice.

## Visual Examples

The example below synced a short video of a person to a speech recording.

- **Input video** (`video_url`): [Watch the input video](https://blog.api.market/wp-content/uploads/2026/10/p-video-avatar-output-api-market.mp4)
- **Input audio** (`audio_file`): [Listen to the input audio](https://blog.api.market/wp-content/uploads/2026/10/realtime-tts-1.5-max-output-api-market.mp3)

Output:

[Watch the example video](https://blog.api.market/wp-content/uploads/2026/10/kling-lip-sync-output-api-market.mp4)

## Pricing

- **Pay As You Go (PAYG) pricing**: You only pay for what you use.
- **Price**: $0.018 per API unit.
- **1 API unit = 1 second of output video**, rounded up to the next whole second (minimum 1 unit).
- **Worked example**: a 5 second video is 5 units, **$0.09**. A 10 second video is 10 units, **$0.18**. A 7.2 second video rounds up to 8 units, **$0.144**.
- **Failed or cancelled predictions**: Not billed.
- **GET requests to check prediction status**: Free and don't count toward API unit usage.
- **No monthly subscription fee**: Pay as you go.

## API Documentation

### Base URL

All API requests go to: `https://prod.api.market/api/v1/magicapi/lip-sync-video-generator-kwaivgi-kling-lip-sync`

### Authentication

Include your api.market API key in the `x-api-market-key` header of every request:

```http
x-api-market-key: YOUR_API_MARKET_KEY
```

### Asynchronous Processing

This API is asynchronous:

1. Send a POST request to `/predictions`. The response is **201 Created** with a prediction `id` and a `status` of `starting` or `processing`. A 201 means the job was accepted; it is not an error.
2. Poll GET `/predictions/{id}` (for example once a second).
3. Stop when `status` is `succeeded` (the result is in `output`), `failed` or `canceled` (details in `error`).

Set a client-side timeout (for example 10 minutes) so a stuck job does not poll forever.

### Endpoints

#### POST /predictions

Creates a prediction. This is the request used for the example above:

```bash
curl -X POST "https://prod.api.market/api/v1/magicapi/lip-sync-video-generator-kwaivgi-kling-lip-sync/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  --data @- <<'EOF'
{
  "version": "8311467f07043d4b3feb44584d2586bfa2fc70203eca612ed26f84d0b55df3ce",
  "input": {
    "video_url": "https://blog.api.market/wp-content/uploads/2026/10/p-video-avatar-output-api-market.mp4",
    "audio_file": "https://blog.api.market/wp-content/uploads/2026/10/realtime-tts-1.5-max-output-api-market.mp3"
  }
}
EOF
```

**Response (201 Created):**

```json
{
  "id": "ufawqhfynnddngldkgtslldrkq",
  "version": "8311467f07043d4b3feb44584d2586bfa2fc70203eca612ed26f84d0b55df3ce",
  "status": "starting",
  "created_at": "2026-10-13T10:45:20.695Z"
}
```

#### GET /predictions/{id}

Returns the prediction's status and, once it has succeeded, its output.

```bash
curl "https://prod.api.market/api/v1/magicapi/lip-sync-video-generator-kwaivgi-kling-lip-sync/predictions/ufawqhfynnddngldkgtslldrkq" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

**Response (200 OK, succeeded):**

```json
{
  "id": "ufawqhfynnddngldkgtslldrkq",
  "status": "succeeded",
  "output": "https://blog.api.market/wp-content/uploads/2026/10/kling-lip-sync-output-api-market.mp4",
  "created_at": "2026-10-13T10:45:20.695Z",
  "started_at": "2026-10-13T10:45:21.102Z",
  "completed_at": "2026-10-13T10:46:05.480Z"
}
```

`output` is a single URL to the lip-synced MP4 video.

### Input Parameters

| Parameter | Type | Default | Required | Description |
|---|---|---|---|---|
| `voice_id` | string | en_AOT | No | Voice ID for speech synthesis (if using text and not audio). See the list of voice IDs below the table. |
| `audio_file` | string (URI) | - | No | Audio file for lip sync. Must be .mp3, .wav, .m4a, or .aac and less than 5MB. |
| `text` | string | - | No | Text content for lip sync (if not using audio) |
| `video_id` | string | - | No | ID of a video generated by Kling. Cannot be used with video_url. |
| `video_url` | string | - | No | URL of a video for lip syncing. It can be an .mp4 or .mov file, should be less than 100MB, with a duration of 2-10 seconds, and a resolution of 720p-1080p (720-1920px dimensions). Cannot be used with video_id. |
| `voice_speed` | number | 1 | No | Speech rate (only used if using text and not audio) Range: 0.8 to 2. |

No field is marked required by the schema, but a request needs one video source (`video_url` or `video_id`, not both) and one speech source (`audio_file`, or `text` with an optional `voice_id` and `voice_speed`). `voice_id` and `voice_speed` are only used with `text`.

**Voice IDs** (`voice_id`): `en_AOT`, `en_oversea_male1`, `en_girlfriend_4_speech02`, `en_chat_0407_5-1`, `en_uk_boy1`, `en_PeppaPig_platform`, `en_ai_huangzhong_712`, `en_calm_story1`, `en_uk_man2`, `en_reader_en_m-v1`, `en_commercial_lady_en_f-v1`, `zh_genshin_vindi2`, `zh_zhinen_xuesheng`, `zh_tiyuxi_xuedi`, `zh_ai_shatang`, `zh_genshin_klee2`, `zh_genshin_kirara`, `zh_ai_kaiya`, `zh_tiexin_nanyou`, `zh_ai_chenjiahao_712`, `zh_girlfriend_1_speech02`, `zh_chat1_female_new-3`, `zh_girlfriend_2_speech02`, `zh_cartoon-boy-07`, `zh_cartoon-girl-01`, `zh_ai_huangyaoshi_712`, `zh_you_pingjing`, `zh_ai_laoguowang_712`, `zh_chengshu_jiejie`, `zh_zhuxi_speech02`, `zh_uk_oldman3`, `zh_laopopo_speech02`, `zh_heainainai_speech02`, `zh_dongbeilaotie_speech02`, `zh_chongqingxiaohuo_speech02`, `zh_chuanmeizi_speech02`, `zh_chaoshandashu_speech02`, `zh_ai_taiwan_man2_speech02`, `zh_xianzhanggui_speech02`, `zh_tianjinjiejie_speech02`, `zh_diyinnansang_DB_CN_M_04-v2`, `zh_yizhipiannan-v1`, `zh_guanxiaofang-v2`, `zh_tianmeixuemei-v1`, `zh_daopianyansang-v1`, `zh_mengwa-v1`.

### Status Codes

- **201 Created**: Prediction accepted (POST). Poll the GET endpoint for the result.
- **200 OK**: Prediction returned (GET).
- **400 Bad Request**: Invalid input, for example a value outside the allowed range or enum.
- **401 Unauthorized**: Missing or invalid `x-api-market-key`.
- **404 Not Found**: Unknown prediction ID.
- **429 Too Many Requests**: Too many requests in a short time. Wait and retry.
- **500 Internal Server Error**: Server-side error. Retry later.

### Error Handling

- Treat both 200 and 201 as success.
- A prediction can still end with `status: "failed"`; read the `error` field for the reason. Failed predictions are not billed.
- Retry 429 and 5xx responses with exponential backoff. Fix the request before retrying a 400.

## Rate Limits

All plans allow 10 requests per second. Requests above that return `429 Too Many Requests`; back off before retrying and honour any `Retry-After` header. Polling a prediction with GET is free but still counts toward the per-second limit, so poll every 1 to 5 seconds rather than in a tight loop.

## Implementation Guide

### JavaScript

```javascript
const apiKey = "YOUR_API_MARKET_KEY";
const baseUrl = "https://prod.api.market/api/v1/magicapi/lip-sync-video-generator-kwaivgi-kling-lip-sync";

async function createAndWaitForPrediction(input) {
  const response = await fetch(`${baseUrl}/predictions`, {
    method: "POST",
    headers: {
      "Content-Type": "application/json",
      "x-api-market-key": apiKey
    },
    body: JSON.stringify({ version: "8311467f07043d4b3feb44584d2586bfa2fc70203eca612ed26f84d0b55df3ce", input })
  });

  // 201 Created is the normal success response here.
  if (!response.ok) {
    throw new Error(`API error: ${response.status} ${await response.text()}`);
  }

  const prediction = await response.json();
  return pollForResult(prediction.id);
}

async function pollForResult(id) {
  const timeoutMs = 10 * 60 * 1000;
  const start = Date.now();

  while (Date.now() - start < timeoutMs) {
    const response = await fetch(`${baseUrl}/predictions/${id}`, {
      headers: { "x-api-market-key": apiKey }
    });
    if (!response.ok) {
      throw new Error(`API error: ${response.status} ${await response.text()}`);
    }

    const prediction = await response.json();
    if (prediction.status === "succeeded") return prediction.output;
    if (prediction.status === "failed" || prediction.status === "canceled") {
      throw new Error(`Prediction ${prediction.status}: ${prediction.error || "unknown error"}`);
    }
    await new Promise((resolve) => setTimeout(resolve, 1000));
  }

  throw new Error("Prediction timed out");
}

createAndWaitForPrediction({
  "video_url": "https://blog.api.market/wp-content/uploads/2026/10/p-video-avatar-output-api-market.mp4",
  "audio_file": "https://blog.api.market/wp-content/uploads/2026/10/realtime-tts-1.5-max-output-api-market.mp3"
}).then(console.log);
```

### Python

```python
import time
import requests

API_KEY = "YOUR_API_MARKET_KEY"
BASE_URL = "https://prod.api.market/api/v1/magicapi/lip-sync-video-generator-kwaivgi-kling-lip-sync"

def create_and_wait_for_prediction(input_data):
    response = requests.post(
        f"{BASE_URL}/predictions",
        headers={"Content-Type": "application/json", "x-api-market-key": API_KEY},
        json={"version": "8311467f07043d4b3feb44584d2586bfa2fc70203eca612ed26f84d0b55df3ce", "input": input_data},
    )
    # 201 Created is the normal success response here.
    if response.status_code not in (200, 201):
        raise Exception(f"API error: {response.status_code} {response.text}")
    return poll_for_result(response.json()["id"])

def poll_for_result(prediction_id, timeout=600):
    start = time.time()
    while time.time() - start < timeout:
        response = requests.get(
            f"{BASE_URL}/predictions/{prediction_id}",
            headers={"x-api-market-key": API_KEY},
        )
        if response.status_code != 200:
            raise Exception(f"API error: {response.status_code} {response.text}")
        prediction = response.json()
        if prediction["status"] == "succeeded":
            return prediction["output"]
        if prediction["status"] in ("failed", "canceled"):
            raise Exception(f"Prediction {prediction['status']}: {prediction.get('error') or 'unknown error'}")
        time.sleep(1)
    raise Exception("Prediction timed out")

print(create_and_wait_for_prediction({
  "video_url": "https://blog.api.market/wp-content/uploads/2026/10/p-video-avatar-output-api-market.mp4",
  "audio_file": "https://blog.api.market/wp-content/uploads/2026/10/realtime-tts-1.5-max-output-api-market.mp3"
}))
```

### cURL

```bash
# Step 1: create the prediction (returns 201 Created with an id)
curl -X POST "https://prod.api.market/api/v1/magicapi/lip-sync-video-generator-kwaivgi-kling-lip-sync/predictions" \
  -H "Content-Type: application/json" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY" \
  --data @- <<'EOF'
{
  "version": "8311467f07043d4b3feb44584d2586bfa2fc70203eca612ed26f84d0b55df3ce",
  "input": {
    "video_url": "https://blog.api.market/wp-content/uploads/2026/10/p-video-avatar-output-api-market.mp4",
    "audio_file": "https://blog.api.market/wp-content/uploads/2026/10/realtime-tts-1.5-max-output-api-market.mp3"
  }
}
EOF

# Step 2: poll with the id from step 1 until status is succeeded, failed or canceled
curl "https://prod.api.market/api/v1/magicapi/lip-sync-video-generator-kwaivgi-kling-lip-sync/predictions/PREDICTION_ID" \
  -H "x-api-market-key: YOUR_API_MARKET_KEY"
```

## FAQ

**Why do I get a 201 response instead of the result?**
The API is asynchronous. 201 Created means the prediction was accepted. Poll GET `/predictions/{id}` until `status` is `succeeded` and read `output`.

**What video can I use?**
An `.mp4` or `.mov` file under 100 MB, 2 to 10 seconds long, with 720 to 1920 px dimensions, passed as `video_url`. Or a video generated by Kling, passed as `video_id`.

**What audio formats are accepted?**
`.mp3`, `.wav`, `.m4a` or `.aac`, under 5 MB.

**Can I lip-sync to text instead of audio?**
Yes. Send `text` and pick a `voice_id` (default `en_AOT`); `voice_speed` sets the speech rate from 0.8 to 2.

**How is a request priced?**
By the length of the output video: $0.018 per second, rounded up to the next whole second.

## Contact Us

If you have any questions or need assistance, you can reach out to us through the embedded chat on the bottom right corner of the page or send an email to hello@magicapi.com.


Sync the lips in a 2 to 10 second video to an audio file or to text read by a built-in voice with Kling Lip Sync. Pay per second of video.