| Field | Type | Required | Example/Default Value | Description | 
|---|---|---|---|---|
| person | object | Yes | Digital human avatar object | |
| ├─ id | string | Yes | C-d1af99e0fee34978bc844d43078bf8c9 | ID returned from the avatar list | 
| ├─ x | int | Yes | 0 | X position | 
| ├─ y | int | Yes | 480 | Y position | 
| ├─ figure_type | string | No | whole_body | Required only when using public digital humans | 
| ├─ width | int | Yes | 1080 | Width | 
| ├─ height | int | Yes | 1920 | Height | 
| ├─ drive_mode | string | No | "" / random | Drive mode, supports normal sequence or random frame action (random), default is normal | 
| ├─ is_rgba_mode | bool | No | false/true | Whether to drive four-channel webm video, default is false | 
| ├─ backway | int | No | 1 | Playback order when driving digital human to the end of material, 1 for forward, 2 for reverse | 
| audio | object | Yes | Audio object | |
| ├─ tts | object | Yes | TTS object | |
| │ ├─ text | string/array | Yes | ["Text content"] | Text, string array, all content in one string, length limit within 4000 | 
| │ ├─ speed | float | No | 1.0 | Speech speed, range 0.5~2 | 
| │ ├─ audio_man | string | Yes | Digital human voice ID | |
| │ ├─ wav_url | string | No | MP3, M4A, or WAV audio file to drive digital human | |
| │ ├─ type | string | Yes | tts / audio | Voice generation type, tts for text-to-speech, audio for audio file | 
| │ ├─ volume | int | No | 100 | Volume | 
| │ ├─ language | string | No | cn | Language type, default is cn | 
| bg_color | string | No | #EDEDED | Background color | 
| bg | object | No | Background image object | |
| ├─ src_url | string | Yes | https://res.chanjing.cc/chanjing/static/res/bg/xxx.png | Background image URL, supports only JPG, PNG formats | 
| ├─ x | int | Yes | 0 | X coordinate | 
| ├─ y | int | Yes | 0 | Y coordinate | 
| ├─ height | int | Yes | 1920 | Image height | 
| ├─ width | int | Yes | 1080 | Image width | 
| subtitle_config | object | No | Subtitle configuration object (optional) | |
| ├─ x | int | No | 31 | Subtitle display range start X coordinate, recommended 31 | 
| ├─ y | int | No | 1521 | Subtitle display range start Y coordinate, recommended 1521 | 
| ├─ show | bool | No | true | Whether to show subtitles | 
| ├─ width | int | No | 1000 | Subtitle display range width, recommended 1000 | 
| ├─ height | int | No | 200 | Subtitle display range height, recommended 200 | 
| ├─ font_size | int | No | 64 | Font size, recommended 64 | 
| ├─ color | string | No | #000000 | Font color value | 
| ├─ stroke_color | string | No | #000000 | Font stroke color value | 
| ├─ stroke_width | int | No | 7 | Font stroke width, recommended 7 | 
| ├─ font_id | string | No | Font ID | |
| screen_width | int | No | 1080 | Screen width, default 1080 | 
| screen_height | int | No | 1920 | 
{
    "person": {
        "id": "a756a5d802064a9a9852ae1742b3499a",
        "x": 0,
        "y": 0,
        "width": 1080,
        "height": 1920,
        "figure_type": "sit_body",
        "drive_mode": "random"
    },
    "audio": {
        "tts": {
            "text": [
                "302AI is truly an excellent platform. At 302, you can not only use almost all AI models available on the market online but also integrate via API, greatly benefiting developers."
            ],
            "speed": 1,
            "audio_man": "C-CASE-56069e0e255c43239cf39f899cfedb42"
        },
        "wav_url": "",
        "type": "tts",
        "volume": 100,
        "language": "cn"
    },
    "bg_color": "#d92127",
    "screen_width": 1080,
    "screen_height": 1920
}curl --location --request POST 'https://api.302.ai/chanjing/open/v1/create_video' \
--header 'Authorization: Bearer ' \
--header 'Content-Type: application/json' \
--data-raw '{
    "person": {
        "id": "a756a5d802064a9a9852ae1742b3499a",
        "x": 0,
        "y": 0,
        "width": 1080,
        "height": 1920,
        "figure_type": "sit_body",
        "drive_mode": "random"
    },
    "audio": {
        "tts": {
            "text": [
                "302AI is truly an excellent platform. At 302, you can not only use almost all AI models available on the market online but also integrate via API, greatly benefiting developers."
            ],
            "speed": 1,
            "audio_man": "C-CASE-56069e0e255c43239cf39f899cfedb42"
        },
        "wav_url": "",
        "type": "tts",
        "volume": 100,
        "language": "cn"
    },
    "bg_color": "#d92127",
    "screen_width": 1080,
    "screen_height": 1920
}'{
    "code": 0,
    "data": "1935639178641408000",
    "msg": "success",
    "trace_id": "b2159e6f8c67e87a891e533e1cef1222"
}