Your flow can have the following types of Transcription tasks: Field Transcription, Table Transcription, and Flexible Extraction.
A Field or Table Transcription task can be generated in the following ways:
The machine's transcription is below the transcription target accuracy thresholds. For more information on how to set thresholds, see Transcription Settings.
If a field has Consensus or the "Supervision Always" setting enabled in the Layout Editor. Note that these tasks have escalated priority.
Flexible Extraction tasks are generated in the following situations:
A Structured page was originally marked as No Layout Found by the machine, but it was later manually matched to a layout in Document Classification Supervision.
The code in a Custom Code Block sent a document to Flexible Extraction so a keyer could verify the content of specific fields.
The code in a Custom Code Block sent a document to Flexible Extraction so a keyer could verify the content of the entire document.
Field Transcription task
The Field Transcription task asks keyers to key the text seen on the page. Depending on the data type, the field may have formatting validations. For example, if the data type is SSN, the UI will show an error message if you try to enter a letter. For more information on this, see What is a Data Type?
The screen that appears as you transcribe a field also provides information on how to mark fields as illegible or N/A, and how to override formatting validations. Note that if you have the Supervision Transcription masking setting enabled in your Manual Transcription Block, you can bypass the formatting restrictions in Field Transcription Supervision (e.g., limitations in characters that can be input) so you do not have to manually override it every time. For more information, see Flow Blocks.
See the screenshot below for reference:
If you’re transcribing a field that has multiple bounding boxes, you need to transcribe all bounding boxes associated with the field. A single task with all bounding boxes appears, and you can navigate between the bounding boxes by clicking the left ( < ) and right ( > ) arrows.
Here are some additional notes about Transcription Supervision:
For Structured documents, you can "Reject" documents to stop processing. For more information, see Configuring Custom Rejection Reasons.
Transcription tasks are generated at the document-level.
There is a 2000-character limit per transcription field. If the content exceeds that limit, we will truncate it to 2000 characters, and we will let you know that we have done so through a field exception. For more information, see Document Output Page.
There is a 200-character limit per line for a transcription field. If the content exceeds that limit, we will truncate it to 200 characters.
For information on supported characters, see Supported Languages and Supported Characters and Default Data Types.
Keyboard shortcuts
Shortcut | Useful tips |
Alt + Ctrl + N | Mark field as N/A. Should be used when someone has filled "N/A", "NA", or "Not Applicable" in the field |
F9 | Override formatting validations. Validations are added to certain data types as specified in the Layout Editor. Use this to transcribe a form field value that hasn’t been defined as a valid value |
F2 | Show/Hide shortcuts guide. The shortcuts guide includes special characters that can be entered, such as international currency (British pound, yen, euro) and MICR characters. |
Table Transcription task
If the document has gone through Table Identification, and at least one table cell was transcribed by the machine at a confidence below the Table threshold, the document will have a Table Transcription step within the Transcription workflow. For nested tables, the document will have a Transcription task only for the table that has at least one low-confidence table cell transcription. If both the child and the parent table have low-confidence cell transcriptions, a single Transcription task will include both tables. You will first need to complete Transcription Supervision for the child table, and then complete Transcription Supervision for the parent table.
Note that Table Identification and Table Transcription tasks will be assigned to the same keyer, whenever possible, to avoid lost context.
Overview
In this step, you can review all the values within the table cells that the machine was able to capture. You can also edit and transcribe any values in the table that the machine may have missed, especially in instances where the text is not precisely aligned within a rectangular cell.
Document panel
The main panel in the center of the screen presents an image of the document pages containing the identified table and all of the fields that require transcription. Those fields will be highlighted in blue boxes.
Transcription work area
Situated under the main panel is the transcription work area. Here, you will type the text in the highlighted fields seen on the page above. Note that if you click an empty field in this area, the main panel will readjust to show you the respective field in the document panel above.
The column headers at the top of the transcription work area will reflect the names of the columns within the table in the main panel.
Interpreting the transcription work area:
Blue borders around a cell indicates that the machine was not sure of its prediction.
An empty cell with a blue border = indicates that the machine was not sure if the cell is indeed empty
A cell with a value and a blue border = indicates that the machine was not sure of the value it predicted
Cells without a border indicate that the machine made a confident prediction
Using the transcription work area:
To edit a cell's content, click within the cell of the spreadsheet.
If the cell was identified during the Table Identification task, the image above the spreadsheet will focus on the cell you have selected and gray out its surrounding contents.
If you would like to match the order of the column headers in the spreadsheet to their order in the document, drag and drop the column headers within the spreadsheet area to reorder them.
Document details panel
To show additional information about the document, click the arrow on the right side of the screen. Here, you will see the Document Details and All Columns dropdown menus.
Document Details – shows the submission number, document number, and layout name
All Columns – shows all of the columns, present or not, and allows you to show/hide their transcription field in the work area
Keyboard shortcuts
Table transcription
Task | Mac Shortcut | Windows Shortcut |
Enter & exit cell | Return | Enter |
Exit cell & discard edits | Esc | Esc |
Next cell | Tab | Tab |
Previous cell | Shift + Tab | Shift + Tab |
Next empty cell | Control + E | Control + E |
Previous empty cell | Shift + Control + E | Shift + Control + E |
Navigate cells | ⬆ or ⬇ or ➡︎ or ⬅︎ | ⬆ or ⬇ or ➡︎ or ⬅︎ |
Override validation | F9 | F9 |
New line | Shift + Return | Shift + Enter |
Add row | Option + N | Alt + N |
Currency symbols
Task | Mac Shortcut | Windows Shortcut |
€ | Option + 3 | Alt + 3 |
$ | Option + 4 | Alt + 4 |
£ | Option + 5 | Alt + 5 |
Â¥ | Option + Y | Alt + Y |
All tasks
Task | Mac Shortcut | Windows Shortcut |
Zoom in | Option + + | Alt + + |
Zoom out | Option + - | Alt + - |
Flexible Extraction task
In Flexible Extraction tasks, keyers are asked to either transcribe the fields they see in a document or to validate the content of fields marked for review. Just as in Field Transcription tasks, each field may have data-type validation rules that prevent you from entering an incorrect type of data.
The system generates Flexible Extraction tasks in the following situations:
When a Structured page is manually matched to a layout, preventing it from going through the typical Transcription Supervision process. In these instances, no fields will be marked as required for review, and the keyer should transcribe whatever fields they see on the page.
When the rules in a Custom Code Block flag a document for further review by a keyer. The rules in the code may require that the entire document be reviewed or only certain fields. The system will mark the fields that the keyer should review, and the keyer won’t be able to complete the task until they’ve entered responses for all these fields.
If a Structured page was matched to the wrong layout before it was sent to Flexible Extraction, you will not be able to assign it to a different layout. However, you can indicate that it was matched to the wrong layout by clicking Mark Layout Variation Incorrect at the top of the right-hand sidebar.
Flexible Extraction tasks are very similar to Field Transcription tasks:
Flexible Extraction tasks are generated at the document-level.
As in Field Transcription is a 2000-character limit per transcription field. If the content exceeds that limit, we will truncate it to 2000 characters, and we will let you know that we have done so through a field exception. For more information, see Document Output Page.
English, French, Spanish, German, Italian, and Dutch characters are accepted.
Fields with multiple occurrences
During Flexible Extraction, you can add, transcribe, and remove fields’ occurrences. All fields’ occurrences that have been identified in previous steps of the flow are displayed in the right-hand sidebar of the Flexible Extraction task’s page.
If you identify additional fields’ occurrences during Flexible Extraction, you can transcribe each occurrence by following the steps below:
Select an occurrence in the right-hand sidebar by clicking on it.
Click on the menu (three dots) next to the occurrence in the right-hand sidebar.
In the drop-down menu, click Add occurrence.
Transcribe the occurrence.
To delete a field’s occurrence, follow the steps below:
Select an occurrence in the right-hand sidebar by clicking on it.
Click on the menu (three dots) next to the occurrence in the right-hand sidebar.
In the drop-down menu, click Remove occurrence.