Using the Personal Asset Image to Create a Stream

Download

Mode fokus

Ukuran font

Terakhir diperbarui: 2026-04-28 15:19:31

It is used to create a new video stream and obtain the playback URL for the digital human video.
Calling Protocol
HTTPS + JSON
POST /v2/ivh/sessionmanager/sessionmanagerservice/createsessionbyasset
Header Content-Type: application/json;charset=utf-8
Request Parameters
Parameter Name
Type
Required
Description
ReqId
String
Yes
Unique identifier for each request, a 32-character UUID.
SessionId
String
No
Unique identifier for the session. Not allowed to be passed in SaaS environments, assigned by the cloud. Can be passed in private deployment environments.
AssetVirtualmanKey
String
Yes
Personal asset image ID. Obtain it from the Asset Management section of the Digital Human platform, as shown in the figure:
﻿
﻿
﻿
UserId
String
Yes
The unique identifier for the user, maintained by the caller. Creating a new stream with the same UserIdentifier will cause the previous stream with that UserIdentifier to close.
Protocol
String
Yes
Currently supported parameters: rtmp, trtc and webrtc
DriverType
int
Yes
The driving method for the digital human.
1: text-only driving;
3: audio driving (original voice), which supports both audio and text driving modes in this setting;
ProtocolOption
ProtocolOption
No
Protocol-specific custom parameters. Ignore if not using trtc.
SpeechParam
SpeechParam
No
Parameters related to timbres.
ExtraInfo
﻿ExtraInfo﻿
No
Extended parameters for stream creation.
Backgrounds
Arrays of [VideoLayer]
No
List of background images. The total number of backgrounds and foregrounds cannot exceed 5.
Note:
This parameter does not take effect for streamers whose shooting background has not been removed.
For streamers whose shooting background is removed, the background application priority is: the background set by this parameter > the green screen.
Foregrounds
Arrays of [VideoLayer]
No
List of foreground images. The total number of backgrounds and foregrounds cannot exceed 5.
AnchorParam
AnchorParam
No
Anchor-related parameters.
﻿ProtocolOption
﻿
Name
Type
Required
Description
TrtcUseExternalApp
boolean
No
It indicates whether to use an external trtc. If you do not use it, the unified trtc appid of the digital human platform will be used.
Notes:
When the unified trtc appid of the digital human platform is used, it is limited to the debugging stage and cannot be used in the production stage.
TrtcAppId
string
No
The trtc appid (required when using an external trtc appid).
TrtcRoomId
int
No
The trtc room ID (The room ID will be assigned by the cloud if it is not specified.).
TrtcUserSig
string
No
The trtc AI Digital Human user signature (required when using an external trtc appid).
TrtcPrivateMapKey
string
No
The trtc AI Digital Human user permission key (required when using an external trtc appid).
CssCustomPushUrl
string
No
Customize a CSS streaming URL. The streaming protocol is fixed to rtmp. Use rtmp to stream to CSS. You can use various playback protocols supported by CSS to pull and play the stream.
The required streaming URL format is:
rtmp://xyz.com/cssAppName/streamid?txSecret={0}&txTime={1}
For the calculation methods of txSecret and txTime, refer to: https://www.tencentcloud.com/document/product/267/32720?from_cn_redirect=1
﻿SpeechParam
﻿
Name
Type
Required
Description
Speed
float
No
Speech speed, with a range of [50, 200]. A value of 50 indicates 50% of the default speed. If the speed is not specified, it is recommended to use the default value of 100.
TimbreKey
string
No
Timbre key.
Volume
int
No
Volume level, with a range of [-10, 10]. A value of -10 indicates a decrease of 10 dB relative to the default volume. If the volume is not specified, it is recommended to use the default value of 0.
TimbreLanguage
string
No
Timbre language. For available languages, refer to the Personal Asset Management API Paginated Query Timbre List interface. When a multilingual timbre is synthesized, you must select the corresponding language.
﻿VideoLayer
﻿
Parameter Name
Type
Required
Description
URL
String
Yes
URL of the layer resource.
Image formats: jpg, jpeg, png, mjpeg, and gif are supported.
Image size: It is recommended not to exceed 2 MB.
X
int
Yes
X-coordinate of the top-left corner for resource placement.
Y
int
Yes
The Y-coordinate of the top-left corner for resource placement.
Width
int
Yes
Width of the resource output.
Height
int
Yes
Height of the resource output.
﻿ExtraInfo
﻿
Name
Type
Required
Description
AlphaChannelEnable
bool
No
Whether to enable the output of the spliced video stream with alpha channel in raw video splicing.
Note:
Assuming the original output width is iW, the actual video stream width is: iW + pW + iW.
pW is the intermediate pad width: (16 - iW%16) % 16.
The starting point of the alpha video width: iW + pW.
The alpha video width: iW.
This feature is not currently supported for the Avatar (High-Precision) model.
﻿AnchorParam
﻿
Parameter Name
Type
Required
Description
SilenceActionDisabled
bool
No
Interactive dialogue mode switch (default: False): Applicable to 2D Avatar Digital Humans. If the avatar material meets the interactive scene recording requirements during recording, you can turn on this switch to enable the digital human to have no additional hand movements during silence, making the interaction more realistic and natural.
Note:
If the avatar itself does not support this feature, this configuration item does not take effect.
Response Parameter
Name
Type
Required
Description
ReqId
String
Yes
Unique identifier for a single request.
SessionId
String
Yes
Unique identifier for the session
SessionStatus
int
Yes
Status:
1: In progress (ready), the cache hit directly provides the playback URL.
3: In progress (not ready). If the cache is missed, wait for the model to load. Use the Query Session Status API to poll the session status until the stream status changes to 1. This process usually takes no more than 2 minutes.
PlayStreamAddr
String
No
The playback URL, in the format: rtmp://liveplay.ivh.qq.com/live/m789
This field will not be returned if a custom CSS streaming URL is specified through the CssCustomPushUrl parameter.
Request Sample
{
    "Header": {
﻿
    },
    "Payload": {
        "ReqId": "d7aa08da33dd4a662ad5be508c5b77cf",
        "AssetVirtualmanKey": "d7aa08da33dd4a662ad5be508c5b77cf",
        "DriverType": 1,
        "UserId": "henry",
        "Protocol": "rtmp"
    }
}
Response Sample
{
    "Header": {
        "Code": 0,
        "Message": "",
        "RequestID": "123"
    },
    "Payload": {
        "ReqId": "d7aa08da33dd4a662ad5be508c5b77cf",
        "SessionId": "m789",
        "SessionStatus": 1,
        "PlayStreamAddr": "rtmp://live.qq.com/live/m789"
    }
}

Bantuan dan Dukungan

Apakah halaman ini membantu?

Anda juga dapat Menghubungi Penjualan atau Mengirimkan Tiket untuk meminta bantuan.

masukan

tencent cloud

Tencent Cloud AI Digital Human

Using the Personal Asset Image to Create a Stream

Calling Protocol

Request Parameters

Response Parameter

Request Sample

Response Sample

Bantuan dan Dukungan

Parameter Name	Type	Required	Description
ReqId	String	Yes	Unique identifier for each request, a 32-character UUID.
SessionId	String	No	Unique identifier for the session. Not allowed to be passed in SaaS environments, assigned by the cloud. Can be passed in private deployment environments.
AssetVirtualmanKey	String	Yes	Personal asset image ID. Obtain it from the Asset Management section of the Digital Human platform, as shown in the figure:
UserId	String	Yes	The unique identifier for the user, maintained by the caller. Creating a new stream with the same UserIdentifier will cause the previous stream with that UserIdentifier to close.
Protocol	String	Yes	Currently supported parameters: rtmp, trtc and webrtc
DriverType	int	Yes	The driving method for the digital human. 1: text-only driving; 3: audio driving (original voice), which supports both audio and text driving modes in this setting;
ProtocolOption	ProtocolOption	No	Protocol-specific custom parameters. Ignore if not using trtc.
SpeechParam	SpeechParam	No	Parameters related to timbres.
ExtraInfo	ExtraInfo	No	Extended parameters for stream creation.
Backgrounds	Arrays of [VideoLayer]	No	List of background images. The total number of backgrounds and foregrounds cannot exceed 5. Note: This parameter does not take effect for streamers whose shooting background has not been removed. For streamers whose shooting background is removed, the background application priority is: the background set by this parameter > the green screen.
Foregrounds	Arrays of [VideoLayer]	No	List of foreground images. The total number of backgrounds and foregrounds cannot exceed 5.
AnchorParam	AnchorParam	No	Anchor-related parameters.

Name	Type	Required	Description
TrtcUseExternalApp	boolean	No	It indicates whether to use an external trtc. If you do not use it, the unified trtc appid of the digital human platform will be used. Notes: When the unified trtc appid of the digital human platform is used, it is limited to the debugging stage and cannot be used in the production stage.
TrtcAppId	string	No	The trtc appid (required when using an external trtc appid).
TrtcRoomId	int	No	The trtc room ID (The room ID will be assigned by the cloud if it is not specified.).
TrtcUserSig	string	No	The trtc AI Digital Human user signature (required when using an external trtc appid).
TrtcPrivateMapKey	string	No	The trtc AI Digital Human user permission key (required when using an external trtc appid).
CssCustomPushUrl	string	No	Customize a CSS streaming URL. The streaming protocol is fixed to rtmp. Use rtmp to stream to CSS. You can use various playback protocols supported by CSS to pull and play the stream. The required streaming URL format is: `rtmp://xyz.com/cssAppName/streamid?txSecret={0}&txTime={1}` For the calculation methods of txSecret and txTime, refer to: `https://www.tencentcloud.com/document/product/267/32720?from_cn_redirect=1`