Convert a PDF to Large Print
Converting a PDF to large print means extracting its text and re-typesetting it at 16, 20 or 24 pt with generous line spacing on clean A4 pages, so a low-vision reader gets comfortable type instead of a magnifier. This page builds that large-print reading edition entirely in your browser, keeping detected headings bold. It is a reflow, not a facsimile: the original layout, images and fonts are not preserved, and scanned PDFs need OCR first.
How it works
Extraction comes first. The tool opens the file with the pdf.js rendering library and reads each page's text items, which arrive as positioned runs — a string plus its transform matrix, from which the run's font size and baseline position can be recovered. Runs are clustered into lines by their vertical coordinates, lines into paragraphs by the gaps between them, and hyphenated line-break fragments are rejoined. This positional reconstruction matters: raw extraction order in a PDF often follows the order objects were written, not reading order, so sorting by geometry is what keeps sentences intact.
Heading detection runs on the collected font sizes. The tool takes the median run size as the document's body size; any line whose size is a clear step above that median — and short enough not to be a pull quote or a paragraph in a big font — is flagged as a heading. Flagged lines survive the conversion in bold with extra space above them, so a 60-page report still reads as chapters and sections rather than one undifferentiated column of text.
Then the re-typesetting. Fresh A4 pages (595.28 × 841.89 pt) are created with wide margins, and the text is set at your chosen size — 16, 20 or 24 pt — with line spacing around 1.5× the type size, the generous leading that large-print publishing standards recommend because tightly packed large type is hardly easier to read than small type. Lines are wrapped by measuring the width of each word run with the font's metrics, and pages break wherever the column fills. The output is honest about what it is: a reading edition. Original fonts, images, columns and tables are gone, which is exactly the trade a large-print reader wants — words first. And because the whole pipeline starts from extractable text, a scanned PDF yields nothing until it has been through OCR.
Illustrative example: a 24-page report at 20 pt
Illustrative scenario, not a recorded benchmark. File sizes, timings and results depend on the input and device. Try the reproducible samples.
A 24-page committee report set at 10.5 pt body type, about 11,200 words, was converted for a reader with macular degeneration. At 20 pt with 30 pt line spacing the reading edition came out at 58 A4 pages — a factor of about 2.4, typical for that jump in size. The 31 section headings, originally 14 and 18 pt, were all detected (each was more than a step above the 10.5 pt median) and appear bold with space above. File size went from 940 KB down to 310 KB, since the reflowed edition carries no images or embedded display fonts. The reader printed it single-sided and read it without a magnifier for the first time; the original stayed alongside for the six charts, which a reflow cannot carry over.
Frequently asked questions
What font size counts as large print?
Accessibility guidance generally puts large print at 16 to 18 points and up, with 20 points or more often called giant print. This tool offers 16, 20 and 24 pt; most regular documents are set at 10 to 12 pt, so even the smallest option is a big step up.
Does the large-print version keep images and layout?
No — this is a reading edition, not a facsimile. The tool extracts the text and re-typesets it on fresh A4 pages, so images, columns, tables and the original fonts are gone. Keep the original file alongside it for figures.
Why does my scanned PDF produce empty large-print pages?
A scanned PDF contains photographs of pages, not extractable text, so there is nothing to re-typeset. Run it through the OCR tool first to recognize the text, then convert the OCRed result to large print.
How many pages will the large-print version have?
Expect roughly 2 to 3 times the original page count at 20 pt, and more at 24 pt, because larger type with wider line spacing fits far fewer words per page. The exact factor depends on how densely the original was set.
How are headings kept in the large-print output?
During extraction the tool records each text run's font size and flags runs noticeably larger than the document's body size as headings. Those lines are set in bold with extra space above them, so the document's structure survives the re-typesetting.