How To Run Text Recognition On Kami

Running text recognition on Kami transforms static PDF documents and scanned image files into fully searchable, selectable text layers using built-in Optical Character Recognition (OCR) technology. By converting flattened pixels into editable character vectors, educators and students can highlight, copy, annotate, and utilize assistive reading tools across digital worksheets with high linguistic accuracy.

Pre-Operation and Scanning Requirements

Transitioning non-searchable digital assets into active, accessible learning materials demands careful attention to source file quality, software subscription tiers, and system configurations. Kami utilizes cloud-based optical processing algorithms that interpret raster images, making source clarity a primary determinant of recognition success.

• Essential Tools & Access: A compatible web browser (Google Chrome or Microsoft Edge), an active internet connection, a Kami account with Teacher or Pro subscription privileges granting access to OCR features, and a target PDF or image file (JPEG, PNG, or PDF).
• Prerequisite Knowledge & Standards: Familiarity with digital file management, understanding of document layers versus flattened image scans, and compliance with institutional data privacy frameworks when processing student documents.
• Estimated Execution Metrics: File processing duration scales linearly with page count, averaging approximately three to seven seconds per page depending on server load, with a recommended maximum single-batch limit of 250 pages for optimal system performance.

Step-by-Step Optical Character Recognition Workflow

Step 1: Access and Import the Target Document

Navigate to the Kami web application interface by logging into your authenticated account dashboard. Click the prominent Upload button located in the upper right quadrant of the screen or select the integration source—such as Google Drive, Microsoft OneDrive, or your local computer storage—to choose the target non-searchable document. Ensure the file has finished rendering completely within the central workspace view before initiating any diagnostic or transformation tasks.

Pro-Tip: If you are uploading physical paper assignments, utilize a mobile scanning application that saves the output directly as a high-contrast PDF with straightened margins to drastically reduce downstream character recognition errors.

Step 2: Open the Document Actions and OCR Toolset

Locate the primary toolbar running vertically or horizontally along the left side of the Kami interface workspace. Click on the Auto Split and Merge icon or navigate directly to the Menu dropdown (represented by three horizontal lines) in the top-left corner. Scroll through the administrative and editing utilities to find the specific tool designated for text recognition or document OCR execution.

Warning: Do not attempt to run text recognition while offline, as Kami requires active connection to cloud processing nodes to execute the neural network character mapping required for modern OCR tasks.

Step 3: Configure Recognition Parameters and Language Settings

Select the target page range for the operation, choosing whether to process the entire document, specific page clusters, or isolated pages containing scanned content. Choose the primary language of the document from the available dropdown menu to ensure the character recognition engine applies the correct linguistic dictionary and phonetic rules, which minimizes character misinterpretation during vectorization.

Step 4: Execute Processing and Verify Text Layer Integration

Click the confirmation button—typically labeled Run OCR or Start Recognition—to initiate the server-side image-to-text conversion algorithm. Monitor the progress bar until the system confirms completion of the document transformation. Test the newly generated text layer by attempting to highlight a sentence using the cursor selection tool or typing a target keyword into the search bar to verify that character vectors are correctly aligned with the underlying visual pixels.

Technical Parameters and Processing Specifications

Common Recognition Failures and Field Fixes

• Root Cause: The scanned document features severe skew, page rotation errors, or heavy shadows along the spine of a bound book.
  
  
  • Actionable Fix: Use image editing software or Kami's built-in crop and rotation tools to straighten the page alignment and increase brightness before running the recognition tool a second time.
• Root Cause: The source file is extremely low resolution (below 150 DPI) or contains heavily pixelated handwriting.
  
  
  • Actionable Fix: Replace the low-fidelity scan with a clean, high-resolution document export, or utilize manual text boxes for handwritten sections since automated OCR struggles with non-standard cursive scripts.
• Root Cause: The selected document language in the OCR configuration menu does not match the actual language present in the text body.
  
  
  • Actionable Fix: Re-open the recognition tool settings, select the exact regional language or dialect matching the document contents, and re-run the batch process to apply the correct character mapping dictionary.

Frequently Asked Questions

What types of files can be processed with Kami text recognition?

Kami supports standard PDF files as well as image formats including JPEG, PNG, and TIFF that have been uploaded into the workspace. If you upload an image format, Kami automatically converts it into a PDF container structure before applying the recognition algorithms.

Why is the text highlighting tool not working after running OCR?

This usually occurs if the recognition process was interrupted or if the document contains complex layered graphics that prevented the software from generating a clean text layer. Try reloading the document in your browser, verifying your internet stability, and running the tool again on the affected pages.

Does Kami OCR support handwritten text recognition?

Kami's automated optical character recognition engine is primarily engineered to process printed, typed, and machine-generated fonts with high precision. While it may occasionally capture clear print-style handwriting, cursive or messy script generally requires manual transcription or annotation using Kami typing tools.

Is there a page limit for running text recognition on a single document?

Yes, processing limits exist to maintain server performance and speed across the cloud platform. While standard multi-page assignments process smoothly, extremely large documents exceeding several hundred pages should be split into smaller batches before initiating recognition.

Can text recognition be run on multiple student documents simultaneously?

Individual file processing is handled per workspace session within the Kami interface. Educators managing large batches of assignments typically process files individually or utilize integrated Learning Management System workflows that support batch document handling.

Master your digital workflow today by transforming static scans into dynamic, fully accessible documents with Kami text recognition.