Description

Example

Click to specify whether you want to use multiple vendors for this project.

NA

Click to specify whether you want labeling analysts to receive mail notifications whenever QA personnel make any changes to tasks on which they had worked. This can serve as a useful feedback loop.

True

Enter email IDs to which notifications must be sent. This field is enabled only if the value in the Enable QC Notification field is True.

john.doe@examplecompany.com

Click to select the percentage of pre-labeled data that must be included into a batch for labeling. This helps you match the labels created by labelers with the expected labels and evaluate the labelers' performance.

10%

Click to specify whether you want to enable a clarification flow, so labelers can ask questions as comments to labels and request guidance from QA team members.

True

Click to specify whether you want to Hide a Name/URL, by default it is Unhide.

NA

Click to specify whether you want to enable project managers to select tasks on which the labelers must work.

For details on selecting tasks in projects, see Managing Project Datasets.

Disabled

Click to specify whether you want to enable named entity recognition in the labeling UI.

True

Click to specify whether you want to enable the labeling UI to allow overlapping polygons during image labeling.

True

Click to specify whether you want to view the labels of the class while annotating .

If set to True: The labels of classes will always be visible.

If set to False: The labels of only the selected class will be visible.

True

Click to specify the annotation tools that must be enabled in the task UI. The available tools are: Rectangle, Polygon, Circle, Ellipse, Cube, Point, Point Group, Polyline, Polyline Group.

Polygon

Click to specify the annotation tools that must be enabled in the task UI. The available tools are: Magic Wand, Four Point, Super Pixel.

Super Pixel

Click to specify whether multiple instances of a segmentation class can be created.

True

Enter the speed (in frames per second) at which you want videos to play by default in the labeling UI.

Note

Taskmonk only supports MP4 and AVI video formats.

24

Enter comma-separated values for different scenes in the video being annotated.

Prologue, Act 1, Act 2, Act 3, Climax

Enter comma-separated values for NER relations that can be captured in a NER Annotation project.

Inferring, Visualizing

Click to select the majority count selection algorithm that you want to use to determine the approved labeling.

Note

This option is enabled only when Execution Method in the Quality Workflow tab is set to Majority Vote.

Majority logic

Click to select the languages in which transcription is required. To enable transliteration of local audio, select the language as English.

English

Click to select the default language used in the transcription text box in the labeling UI. By default, this language is English.

English

Click to select the secondary language that can be used in the transcription text box in the labeling UI. By default, this language is English. If you choose a language here, then a Language field will be added to the labeling UI. By default, this field will be set to the default language chosen above. Labelers will have the ability to click and select the secondary language as required.

Hindi

Click to specify whether you want to enable the labeler to split the transcription into multiple segments. This will enable the labeler to transcribe the contents of specific segments in the audio file into separate text boxes.

True

Click to create tags that can be used to categorize filler words and phrases that are removed from the transcription to make the information clearer.

To manage filler tags:

1. Click Manage Tags and Labels. The Manage Tags popup appears.

2. Enter the name of the tag in the Tag field.

3. Enter a label for the tag in the Label field. This is the label that will appear in the labeling UI.

4. Click Tag Type and select the following tags:

   • Normal: Use this tag to categorize normal sounds that people make during discussions, such as clearing of the throat, laughing, clapping, and so on.

   • Single: Use this tag to categorize single words that are part of the audio file but do not add any value to the discussion. For example, the phrase “Yes, that’s a good idea” would be transcribed as “That’s a good idea”, and “Yes” would be tagged as Single.

   • Filler: Use this tag to categorize onomatopoeic fillers, such as Umm, Ah, and so on, that typically appear in a discussion.

5. Click Save and Add Another to add the filler to the Added Tags and Labels list and create a new one.

6. Click Save and Close once you are done adding tags.