How to Add Text, Edit Content & Scan PDFs in Utilvo
This guide covers every text-related tool in the Advanced PDF Editor: the Text annotation, the Text Callout, the OCR Scan report, the Find & Replace panel, and the practical Whiteout + Text overlay pattern that most readers use to fix typos in real documents. All of it runs client-side; no upload is required.
Utilvo does not run Tesseract.js. See Figure 5 for what OCR Scan actually reports.
Where the Text Tools Live
Unlike most editors that group text tools into one tab, Utilvo splits text work across three groups by intent:
Contains the Text annotation and Text Callout tools. These drop new text on the page.
Contains the OCR Scan tool. Reports which pages have extractable text.
Magnifier icon opens the Find & Replace panel.
Add a Text Annotation
The Text tool (T) places an editable text box anywhere on the page. It's the right tool for adding marginalia — reviewer notes, dates, signatures of approval, or new headings that didn't exist in the source document.
Each Text annotation supports the same typography controls as a desktop publishing tool: font family (Helvetica, Times, Courier, and a dozen more), font size (6-96 pt), color, opacity, alignment, and line height. Bold, italic, and underline are available via the floating context toolbar that appears when you select the annotation.
- Click Annotate → Text in the toolbar (or press T).
- Click anywhere on the page — a text box appears at that location.
- Type your text. The box auto-grows vertically; resize horizontally by dragging the right edge.
- Select the annotation and use the context toolbar to change font, size, color, or alignment.
- Click outside the box to commit the change.
Text Callouts
The Text Callout is a Text annotation with a tail pointing to a specific region of the page. It's the right tool when you want to comment on a specific clause, image, or data point without obscuring it. The tail position is set on click; the text body can be dragged independently afterward.
The Whiteout + Text Overlay Pattern
Utilvo does not modify the PDF's content stream directly — there is no "click a typo and replace it" inline editor. Instead, the practical workflow for editing an existing word or phrase is a two-step overlay:
- Switch to the Whiteout tool (Annotate → Whiteout) and drag a white rectangle over the text you want to replace.
- Switch to the Text tool, click inside the white rectangle, and type the new text.
The Whiteout annotation draws an opaque white rectangle that hides the original content during export. The new Text annotation sits on top, so the reader sees the corrected version while the underlying glyph operators remain in the file. This pattern works for fixing a date, replacing a name, or correcting a number in an invoice.
Whiteout is white, Redact is black. Use Whiteout when you plan to overlay new content; use Redact when you want the covered text gone for good (a permanent-looking black box). Both annotations are exported as filled rectangles drawn on top of the page content.
Find & Replace
The Find & Replace panel opens from the magnifier icon in the top bar. It searches annotation text only — Sticky Note bodies, Text annotations, Stamps, and Callouts — not the PDF's underlying text layer. This is intentional: most PDFs don't have a clean text layer (they're scanned or imaged), and searching across every annotation is what reviewers actually want.
To search the PDF's own text layer (when it exists), use the OCR Scan report (below) to confirm the pages contain extractable text, then use your browser's built-in Ctrl+F / ⌘F shortcut on the rendered page.
OCR Scan: What It Actually Does
The OCR Scan tool does not run an OCR engine. Despite the icon and label, clicking it does not call Tesseract or any other OCR model. Instead, it uses pdf.js's text-extraction API to report which pages already contain extractable text and which are image-only. The result is an alert dialog you can act on.
A usable Tesseract bundle adds ~15 MB to the editor's initial download and pushes memory use above 1 GB while running. That cost would be paid by every user on every page load, even those who never scan a document. Utilvo chose to keep the editor small and fast instead.
Recommended workflow for scanned PDFs: Use the OCR Scan tool to identify which pages are image-only. Export those pages (or the whole document) and run them through a dedicated OCR tool — Adobe Acrobat's OCR, ABBYY FineReader, or a free alternative like tesseract-ocr's command line. Then re-upload the searchable PDF to Utilvo for editing.
Text Tool Quick Reference
| Tool | Group | What it does | Best for |
|---|---|---|---|
| Text | Annotate | Editable text box on the page | Adding marginalia, dates, headings |
| Text Callout | Annotate | Text box with a tail pointing at content | Commenting on a clause or image |
| Whiteout | Annotate | Opaque white rectangle | Covering text before overlay |
| Redact | Annotate / Marks | Opaque black rectangle | Hiding PII permanently |
| Find & Replace | Top bar | Searches annotation text | Reviewing your own notes |
| OCR Scan | Marks | Reports text-presence per page | Diagnosing scanned PDFs |
Further Reading
- Wikipedia — Optical character recognition— what OCR is, how it works, and accuracy tradeoffs.
- Wikipedia — Tesseract (software)— the open-source OCR engine that powers most modern desktop OCR.
- Wikipedia — PDF/A— standard for long-term document preservation and searchable text layers.
Frequently Asked Questions
What OCR engine is used?
Utilvo runs Tesseract.js (a WebAssembly build of Tesseract) directly in your browser. The same engine that powers Google Docs and many open-source OCR pipelines.
Which languages are supported?
Tesseract's full language pack library is available — English, Arabic, French, Spanish, German, Chinese, Japanese, Korean, plus 100+ others.
Can I edit the OCR result?
Yes. After OCR finishes, switch to the Edit Text tool. Every recognized word is rendered as an inline editable node.
Is OCR performed online?
No. The OCR computation, language model download, and text layer assembly all execute inside your browser tab. The original scan is never sent to a server.
Will the OCR text replace the original image?
No. The OCR produces an invisible text layer behind the page image. The page still looks identical, but the text becomes selectable, searchable, and accessible to screen readers.
Last Group: Protect & Sign
Finish the editor tour with password protection, redaction, and signatures.
Read: Protect & Sign PDF