Querying Details About a Virtual Avatar Model Training Task

project_id

Yes

String

Project ID. For details about how to obtain the project ID, see Obtaining a Project ID.

job_id

Yes

String

Task ID.

X-Auth-Token

No

String

User token. This parameter is mandatory when token authentication is used.

You can obtain the token by calling the IAM API used to obtain a user token.

Value of X-Subject-Token in the response header.

Authorization

No

String

Authentication information. This parameter is mandatory for AK/SK authentication.

X-Sdk-Date

No

String

Time when the request is sent. This parameter is mandatory for AK/SK authentication.

The format is YYYYMMDD'T'HHMMSS'Z'.

X-Project-Id

No

String

Project ID. This parameter is mandatory for AK/SK authentication.

X-App-UserId

No

String

Third-party user ID, which does not allow Chinese characters.

X-Request-Id

String

Request ID.

job_id

String

Task ID.

name

String

Virtual avatar model name, which is used as its name in the asset library.

state

String

Task status.

Mapping between these statuses and the statuses displayed on the MetaStudio console:

• To be submitted

  • WAIT_FILE_UPLOAD: to upload files

• Auto reviewing...

  • AUTO_VERIFYING: auto reviewing in progress

  • MANUAL_VERIFYING: manually reviewing in progress

• Failed the auto review

  • AUTO_VERIFY_FAILED: auto review failed

  • MANUAL_VERIFY_FAILED: manual review failed

• Algorithm training ongoing

• MANUAL_VERIFY_SUCCESS: approved and waiting for resource preprocessing

• WAIT_TRAINING_DATA_PREPROCESS: waiting for training data preprocessing

• TRAINING_DATA_PREPROCESSING: training data being preprocessed

• TRAINING_DATA_PREPROCESS_FAILED: training data preprocessing failed

• TRAINING_DATA_PREPROCESS_SUCCESS: training data preprocessed and waiting for resource training

• TRAINING: training...

• TRAIN_FAILED: training failed

• TRAIN_SUCCESS: training completed and waiting for resource preprocessing

• INFERENCE_DATA_PREPROCESSING: inference data being preprocessed

• INFERENCE_DATA_PREPROCESS_FAILED: inference data preprocessing failed

• WAIT_MAIN_FILE_UPLOAD: waiting for main file upload

• MANUAL_STOP_INFERENCE_DATA_PREPROCESS: inference preprocessing manually stopped

• MANUAL_STOP_TRAIN: training manually stopped

• MANUAL_STOP_TRAINING_DATA_PREPROCESS: training preprocessing manually stopped

• WAIT_ADMIN_CONFIRM: waiting for review by administrator

• WAIT_COMPILE: waiting for compilation

• COMPILING: compiling in progress

• COMPILE_FAILED: compilation failed

• WAIT_GENERATE_ACTION: waiting for generating an atomic action

• WAIT_ARRANGE: waiting for choreography

• ACTION_GENERATE_DATA_PROCESSING: an atomic action being generated

• MANUAL_STOP_ACTION_GENERATE_DATA_PROCESSING: action generation manually stopped

• MANUAL_STOP_ACTION_GENERATE_ORI_PROCESSING: choreography manually stopped

• ACTION_GENERATE_ORI_PROCESSING: choreographing in progress

• ACTION_GENERATE_DATA_FAILED: action generation failed

• ACTION_GENERATE_ORI_FAILED: generating the choreography asset failed

• ACTION_GENERATE_ORI_SUCCESS: choreography succeeded

• **GENERATE_ACTION_PREPROCESS_FAILED: generating the atomic action for choreography failed

• WAIT_ADMIN_CALIBRATION: waiting for the administrator to confirm the action

• WAIT_ASSET_SYNC: waiting for asset synchronization

• WAIT_GENERATE_ACTION_MARK: waiting for generating an action mark

• ACTION_MARKING: action marking generated

• To be reviewed by users. This state is available only to whitelisted NAs.

  • WAIT_USER_CONFIRM: waiting for the user to confirm the training result

• Rejected by users. This state is available only to whitelisted NAs.

  • JOB_REJECT: task rejected

• Completed

  • JOB_SUCCESS: training task completed. In this case, the model is available for the user.

  • JOB_FINISH: task ended. This is the final state and cannot be changed. (This state is visible only for NA users, indicating that the model has been accepted by the user.)

• Suspended. This state is available only to whitelisted NAs.

  • JOB_PENDING: task suspended

• WAIT_TEST_VIDEO_CHECK: waiting for test video inference

• TEST_VIDEO_CHECK_PROCESSING: testing video inference under quality check

• TEST_VIDEO_CHECK_SUCCESS: quality check on test video inference passed

• TEST_VIDEO_CHECK_FAILED: quality check on test video inference failed

• VIDEO_ANALYZE_PROCESSING: video check in progress

• VIDEO_ANALYZE_SUCCESS: video check passed

• VIDEO_ANALYZE_FAILED: video check failed

• ACTION_MARKING: action marking in progress

• ACTION_MARK_SUCCESS: action marking succeeded

• ACTION_MARK_FAILED: action marking failed

• ACTION_MARK_UPLOADED: action marking file uploaded

• MANUL_STOP_ACTION_MARK: action marking stopped

asset_id

String

Model asset ID.

project_id

String

Project ID of a model asset.

cover_download_url

String

URL for downloading the virtual avatar model thumbnail. The URL is valid for 24 hours.

last_update_time

String

Indicates the last time when a user updates a task (for example, creates or resubmits a task). The format complies with RFC 3339, for example, 2020-07-30T10:43:17Z.

create_time

String

Creation time. The format complies with RFC 3339, for example, 2020-07-30T10:43:17Z.

contact

String

Mobile number of the creator of the virtual avatar training task.

batch_name

String

Batch name of virtual avatar training tasks.

tags

Array of strings

Tag of a virtual avatar training task.

model_version

String

Virtual avatar model version. The default version is V3.2.

• V3.2: V3.2 model

• V3.3: training model for the ultra-fast Flexus edition

Value range:

0~64

Default value:

V3.2

matting_type

String

Matting type. The default value is AI.

• AI: AI matting

• MANUAL: manual matting

Default value:

AI

model_resolution

String

Virtual avatar model resolution. The default value is 1080P.

• 1080P: supports 1080P and 720P video output.

• 4K: supports 4K, 1080P, and 720P video output.

Default value:

1080P

app_user_id

String

Customized user ID. This parameter is carried if X-App-UserId is set during task creation.

is_flexus

Boolean

Whether the image training is of the basic edition

is_live_copy

Boolean

Whether it is a task for training a livestream-based avatar.

is_fast_flexus

Boolean

Whether it is an ultra-fast Flexus edition.

Default value:

false

is_only_human_model

Boolean

Whether to train the image model but not the voice model. This parameter is valid only for Flexus. The default value is false.

Default value:

false

optional_training_location

Array of strings

Optional training region.

is_background_replacement

Boolean

Indicates whether to replace the background for a virtual avatar. If yes, training videos require green-screen shooting.

Default value:

true

is_ondemand_resource

Boolean

Whether the task is on demand.

Default value:

false

training_video_download_url

String

URL for downloading the virtual avatar training video. The URL is valid for 24 hours.

id_card_image1_download_url

String

URL for downloading the ID card front-side photo. The URL is valid for 24 hours.

id_card_image2_download_url

String

URL for downloading the ID card back-side photo. The URL is valid for 24 hours.

grant_file_download_url

String

URL for downloading the authorization letter. The URL is valid for 24 hours.

pre_beauty_image_download_url

String

URL for downloading the pre-brightening image.

external_action_json_data_download_url

String

URL for downloading the authorization letter. It is valid within 24 hours.

action_video_download_url

String

Action video

post_beauty_image_download_url

String

URL for downloading the brightened image.

audio_file_download_url

String

URL for downloading an audio file.

action_mark_file_download_url

String

URL for downloading the action marking file.

action_mark_file_upload_url

String

URL for uploading the action marking file.

operation_logs

Array of OperationLogInfo objects

Operation logs.

verify_video_matting_info

Array of VerifyVideoMattingInfo objects

Non-cropped area when the crop verification video is being generated.

comment_logs

Array of CommentLogInfo objects

Comment logs.

samples

Array of ActionSampleInfo objects

Action video sample.

is_mask_file_uploaded

Boolean

Indicates whether the mask file is uploaded.

mask_file_download_url

String

URL for downloading the mask. The URL is valid for 24 hours.

verify_video_download_url

String

URL for downloading the video to be reviewed.

markable_video_download_url

String

URL for downloading the tagged video. The URL is valid for 24 hours.

markable_action_video_download_url

String

URL for downloading the action marking video. It is valid within 24 hours.

traning_video_mark_info

TrainingVideoMarkInfo object

Tag information of the virtual avatar training video.

inference_data_process_video_mark_info

InferenceVideoMarkInfo object

Information of virtual avatar inference preprocessing video tag.

voice_properties

VoiceProperties object

Information of virtual avatar inference preprocessing video tag.

inference_data_process_action_mark_info

InferenceActionMarkInfo object

Information of virtual avatar inference preprocessing action tag.

inference_data_process_chat_action_mark_info

InferenceActionMarkInfo object

Tag information of inference preprocessing of actions performed by an interactive virtual avatar.

inference_data_process_eye_correction_mark_info

InferenceEyeCorrectionMarkInfo object

Information of virtual avatar inference preprocessing video tag.

worker_type

Array of strings

Compilation task model.

voice_train_job_id

String

Voice training task ID.

flexus_retry_count

Integer

Number of remaining retraining times of a Flexus task. The value decreases by 1 each time the task is retrained. When the value decreases to 0, the task cannot be retrained.

Value range:

0~10

audio_source_type

String

Audio source.

• VIDEO: audio extracted from a video

• AUDIO: audio uploaded separately

supported_service

Array of strings

Service types supported by the model generated by the task. You can select multiple service types.

Flexus virtual avatars support only VIDEO_2D.

Default value:

[VIDEO_2D,LIVE_2D]

allocated_resource

TrainingAllocatedResource object

Resources allocated for avatar training.

train_video_rotation_state

String

Status of the training video rotation task.

• WAITING: waiting for processing

• PROCESSING: being processed

• SUCCESS: task processing succeeded

• FAILED: task processing failed

action_video_rotation_state

String

Status of the training video rotation task.

• WAITING: waiting for processing

• PROCESSING: being processed

• SUCCESS: task processing succeeded

• FAILED: task processing failed

sub_training_job_info

Array of SubTrainingJobInfoDto objects

Subtask information.

operate_time

String

Operation time. The format complies with RFC 3339, for example, 2020-07-30T10:43:17Z.

log_type

String

Command output.

• USER_CREATE_JOD: Virtual avatar customization started.

• USER_VERIFYING_SUBMITTED: Video submitted for review.

• SYSTEM_VERIFY_FAILED: Auto review failed.

• ADMIN_UPDATE_BEAUTY_LEVEL: Brightening level updated by the administrator.

• ADMIN_UPDATE_JOB_PRIORITY: Task priority updated by the administrator.

• SYSTEM_VERIFY_SUCCESS: Auto review succeeded.

• ADMIN_VERIFY_SUCCESS: Manually approved.

• ADMIN_VERIFY_FAILED: Manual review failed.

• SYSTEM_TRAIN_DATA_PREPROCESSING: Training data being preprocessed.

• SYSTEM_TRAIN_DATA_PREPROCESS_FAILED: Training data preprocessing failed.

• SYSTEM_TRAIN_DATA_PREPROCESS_SUCCESS: Training data preprocessed.

• SYSTEM_ACTION_GENERATE_DATA_PREPROCESSING: Atomic action for choreography being generated.

• SYSTEM_ACTION_GENERATE_DATA_SUCCESS: Atomic action for choreography generated.

• SYSTEM_ACTION_GENERATE_ORI_SUCCESS: Choreography asset data generated.

• SYSTEM_ACTION_GENERATE_DATA_FAILED: Generating the atomic action for choreography failed.

• SYSTEM_ACTION_GENERATE_ORI_FAILED: Generating the choreography asset data failed.

• SYSTEM_ACTION_GENERATE_ORI_PREPROCESSING: Choreography asset data being generated.

• SYSTEM_TRAINING: Training started.

• ADMIN_STOP_TRAINING_DATA_PREPROCESS: Training manually stopped.

• ADMIN_STOP_BEAUTY_PREPROCESS: Brightening manually stopped.

• ADMIN_STOP_INFERENCE_DATA_PREPROCESS: Inference preprocessing manually stopped.

• ADMIN_STOP_TRAIN: Training manually stopped.

• SYSTEM_TRAIN_FAILED: Training failed.

• SYSTEM_TRAIN_SUCCESS: Training succeeded.

• SYSTEM_INFERENCE_DATA_PREPROCESSING: Inference data being preprocessed.

• SYSTEM_INFERENCE_DATA_PREPROCESS_FAILED: Inference data preprocessing failed.

• SYSTEM_INFERENCE_DATA_PREPROCESS_SUCCESS: Inference data preprocessed.

• SYSTEM_JOB_SUCCESS: Task completed.

• ADMIN_MASK_UPLOADED: Mask file uploaded (deprecated).

• ADMIN_UPDATE_VIDEO: Video replacement by administrator.

• ADMIN_UPDATE_ACTION_VIDEO: Video for choreography replaced by administrator.

• ADMIN_RESET: One-click resetting by administrator.

• ADMIN_ACCEPT: Approved by administrator.

• USER_REPAIR: Repaired by users.

• SYSTEM_UPDATE_COVER: Thumbnail changed.

• SYSTEM_ANALYSE_FILE_INFO: Parsing file information.

• ADMIN_SET_SILENCE_REPEAT_NUM: Number of silent rounds set by administrator.

• SYSTEM_MARKABLE_VIDEO: Generating a labeling video.

• SYSTEM_MASK_VERIFY_VIDEO: Generating a verification video.

• SYSTEM_MASK_VERIFY_VIDEO_SUCCESS: Verification video generated.

• SYSTEM_MASK_VERIFY_VIDEO_FAILED: Verification video generation failed.

• SYSTEM_MARKABLE_VIDEO_SUCCESS: Labeling video generated.

• SYSTEM_BEAUTY_PREPROCESSING: Brightening in progress.

• SYSTEM_BEAUTY_PREPROCESS_FAILED: Brightening failed.

• ADMIN_CONFIRM_ACTION: The administrator confirms the action.

• ADMIN_STOP_ACTION_GENERATE_DATA_PREPROCESS: Atomic action generation manually stopped.

• ADMIN_STOP_ACTION_GENERATE_ORI_PREPROCESS: Choreography manually stopped.

• SYSTEM_BEAUTY_PREPROCESS_SUCCESS: Brightening video training data preprocessed.

• SYSTEM_COMPILE_FAILED: Compilation failed.

• SYSTEM_COMPILE_SUCCESS: Compilation succeeded.

• SYSTEM_MARKABLE_VIDEO_FAILED: Labeling video generation failed.

• ADMIN_UPDATE_COMPILE: The administrator updates the compilation configuration.

• ADMIN_UPDATE_INFERENCE_DATA_PROCESS_VIDEO: The administrator updates the time segment information of the inference preprocessing.

• SYSTEM_EXECUTE_COMPILE: Compilation executed.

• SYSTEM_EXECUTE_BEAUTY: Brightening executed.

• SYSTEM_MASK_VIDEO_AND_ACTION_TIME_SUCCESS: Auto labeled.

• SYSTEM_MASK_VIDEO_AND_ACTION_TIME_FAILED: Auto labeling failed.

• USER_UPDATE_VIDEO: Video update by the user.

• ADMIN_UPLOAD_JSON_DATA: Action data uploaded by the administrator.

• ADMIN_DELETE_JSON_DATA: Action data deleted by the administrator.

• ADMIN_UPDATE_GENERAL_CONFIG: Common configuration updated by the administrator.

• ADMIN_MASK_ACTION_TIME: Labeling by the administrator.

• STOP_COMPILE: Compilation manually stopped.

• MAKE_TEST_VIDEO: Marked as a test video.

• ADMIN_SET_FLEXUS_RETRY_COUNT: Number of Flexus retry times set by the administrator.

• USER_DELETE_JOB_VIDEO: Deleting task-related videos by the user.

• ADMIN_SET_VIDEO_ROTATION_ANGLE: Setting the rotation angle of the video by the administrator.

• ADMIN_RE_SET_VIDEO_ROTATION_ANGLE: Restoring the rotation angle of the video by the administrator.

• SYSTEM_SET_VIDEO_ROTATION_ANGLE_SUCCESS: Video rotation succeeded.

• SYSTEM_SET_VIDEO_ROTATION_ANGLE_FAILED: Video rotation failed.

• COMPILE_JOB_IS_CONSUME: Compilation task consumed.

• RESTART_TEST_VIDEO_CHECK: Re-executing the test video check task.

• SKIP_TEST_VIDEO_CHECK: Skipping the test video check task.

• WAIT_TEST_VIDEO_CHECK: Waiting for the test video check task.

• TEST_VIDEO_CHECK_PROCESSING: Testing video check in progress.

• TEST_VIDEO_CHECK_SUCCESS: Test video check succeeded.

• TEST_VIDEO_CHECK_FAILED: Test video check failed.

• REDO_INFERENCE_PREPROCESSING: Inference data preprocessing restarted.

• REDO_TRAINING_PREPROCESSING: Training data preprocessing restarted.

• REDO_TRAINING: Training restarted.

• REDO_ACTION_DATA_GENERATE: Regenerating an atomic action.

• REDO_ACTION_ORI_GENERATE: Action choreography restarted.

• VIDEO_ANALYZE_PROCESSING: Video check in progress.

• VIDEO_ANALYZE_SUCCESS: Video check passed.

• VIDEO_ANALYZE_FAILED: Video check failed.

• ADMIN_RESOLUTION_NORMALIZE: Normalizing the resolution by the administrator.

• SYSTEM_SET_RESOLUTION_NORMALIZE_SUCCESS: Resolution normalized by the administrator.

• SYSTEM_SET_RESOLUTION_NORMALIZE_FAILED: Failed to normalize the resolution by the administrator.

• SYSTEM_ACTION_MARK_PREPROCESS_FAILED: Action marking failed.

• SYSTEM_ACTION_MARK_PREPROCESSING: Action marking is being generated.

• SYSTEM_ACTION_MARK_PREPROCESS_SUCCESS: Action marking succeeded.

• REDO_ACTION_MARK: Regenerating an atomic action mark.

• CONFIRM_ACTION_MARK: Confirming atomic action re-marking.

• MANUL_STOP_ACTION_MARK: Action marking stopped.

• TIME_OUT_RETRY: Retrying upon timeout.

log_description

String

Log description. It is used to record the review comments when the manual or automatic review fails.

operate_user

String

Operators. Options:

• USER

• ADMIN

• SYSTEM

error_code

String

Error code when a request is automatically rejected:

• 44005901: Unsupported video resolution. Only 1080p (1920 x 1080 or 1080 x 1920) and 4K (3840 x 2160 or 2160 x 3840) are supported.

redo_reasons

Array of strings

List of reasons why the task is re-executed by the administrator.

dx

Integer

X-axis position of the pixel in the upper left corner of the area.

Value range:

0~3840

dy

Integer

Y-axis position of the pixel in the upper left corner of the area.

Value range:

0~3840

width

Integer

Area width (pixels).

Value range:

1~7680

height

Integer

Area height (pixels).

Value range:

1~7680

method

String

Resource operation. Options:

• RESERVED: video area that does not need to be cropped

• DELETE: video area that needs to be cropped

• NO_DEGREEN: video area that does not need de-greening

operate_time

String

Operation time. The format complies with RFC 3339, for example, 2020-07-30T10:43:17Z.

comment_type

String

• USER_REJECT: Rejection by user.

• USER_COMMENT: User comments.

• ADMIN_COMMENT: Administrator's reply.

comment_title

String

Title.

comment_message

String

Message.

attachment_download_url

Array of strings

URL for downloading the attachment.

action_name_zh

String

Atomic action name in Chinese.

action_name_en

String

Atomic action name in English.

action_tag

String

Action tag.

catalog

String

Action category name.

recommended_value

Integer

Recommendation level.

Value range:

0~100

is_selected

Boolean

Whether to select this action.

Default value:

true

sample_download_url

String

URL for downloading the atomic action sample file. Valid within 24 hours.

video_start_time

String

Start time of the training video. Format: HH:MM:SS.mmm.

video_end_time

String

End time of the training video. Format: HH:MM:SS.mmm.

video_start_time

String

Start time of the inference preprocessing video. Format: HH:MM:SS.mmm.

video_end_time

String

End time of the inference preprocessing video. Format: HH:MM:SS.mmm.

chat_video_start_time

String

Start time of the interactive video for inference data preprocessing. Format: HH:MM:SS.mmm.

chat_video_end_time

String

End time of the interactive video for inference data preprocessing. Format: HH:MM:SS.mmm.

job_tag

String

Task tag, which is a configuration item of the Flexus edition.

• ECOMMERCE

• NEWS

• MARKETING

sex

String

Voice gender, which is a configuration item of the Flexus edition.

• FEMALE

• MALE

Default value:

FEMALE

language

String

Training language, which is a configuration item of the Flexus edition. Currently, only Chinese is supported.

• CN: Chinese

• EN: English

action_info

Array of ActionMarkItem objects

Actions.

action_start_time

String

Start time of the inference preprocessing video. Format: HH:MM:SS.mmm.

action_end_time

String

End time of the inference preprocessing video. Format: HH:MM:SS.mmm.

action_type

String

Action type. Options:

• SILENCE

• ACTION

Value range:

0~64

eye_correction_start_time

String

Select the start time of inference preprocessing for eye contact correction. Format: HH:MM:SS.mmm.

eye_correction_end_time

String

Select the end time of inference preprocessing for eye contact correction. Format: HH:MM:SS.mmm.

resource_id

String

Resource ID.

charge_mode

String

Resource billing type.

• PERIODIC: yearly/monthly billing

• ONE_TIME: one-off billing

• ON_DEMAND: pay-per-use billing

expire_time

String

Resource expiration time. The format complies with RFC 3339, for example, 2021-01-10T08:43:17Z.

job_id

String

Task ID.

sub_job_type

String

Subtask type.

heart_beat_id

String

Subtask heartbeat ID.

state

String

Task status.

create_time

Long

Creation time.

last_update_time

Long

Last update time.

error_code

String

Error code.

error_msg

String

Error description.

error_code

String

Error code.

error_msg

String

Error description.

200

The information is returned when the request succeeded.

400

Parameters error, including the error code and its description.

500

Internal service error.