Google Drive OCR: how to turn a PDF or photo into text

How to OCR a file with Google Drive

1. Upload the file to Drive. A PDF, or a JPEG, PNG or GIF photo, of 2 MB or smaller[1].
2. Right click it and choose Open with Google Docs. Google Drive detects the language of the document by itself[1].
3. Wait for the new Doc. Drive converts the file and opens the recognized text as a Google Doc[1].
4. Copy or edit the text. Check the layout first, because the format might not transfer from the original[1].

What Google Docs OCR keeps and what it drops

Google's own help page is clear about the format. Bold, italics, font size, font type and line breaks are likely to be retained. Lists, tables, columns, footnotes and endnotes are not likely to be detected[1].

File types

PDF (multipage documents) and JPEG, PNG or GIF photos[1]

File size

2 MB or smaller[1]

Text size

At least 10 pixels high[1]

Orientation

Right side up. Rotate the image before you upload it[1]

Fonts

Common fonts such as Arial or Times New Roman work best[1]

Languages

Detected automatically, from a long list that includes Arabic, Hebrew, Korean and Russian[1]

For a letter or a page of running text, that is enough. For a form, a transcript, an invoice or a certificate, the meaning sits in the table. When the table comes back as loose lines, you rebuild it by hand.

Google Drive OCR and the free OCR on this page

You need no Google account. Upload the image above, press Get the text, and copy the result or download it.

Why does Google Drive OCR fail on my file?

Most failures come from the file, not from Drive. Google lists what it expects[1], and a file that misses one of these points often comes back as an image with little or no text under it.

• The file is bigger than 2 MB. Phone photos are often larger, so shrink the photo or crop it to the page[1].
• The text is small. Google asks for text at least 10 pixels high, so fill the frame with the page[1].
• The page is sideways or upside down. Rotate it first[1].
• The photo is an iPhone HEIC file. Drive's OCR takes JPEG, PNG and GIF photos[1]. The LoreAI box above reads HEIC directly.

A scanned document of many pages

For a whole scanned document, use the LoreAI app. It reads every page of a PDF, a JPG or a PNG with OCR, stamps and handwriting included, and returns one editable Word file with headings, lists and tables in the original order. Your first document is free, up to 3 pages and 3 MB.

The app can also translate the document in the same step into one of 105 languages, with the source language detected automatically. For a scan that only needs to become Word, read OCR PDF to Word.

Questions people ask