How to adjust the formatting of a PDF that is out of format after translation

Key points of this article

For local problems such as line breaks in paragraphs and overflow of individual cells, it is faster to directly change the paragraph text in the PDF editor than to re-translate the entire text.
The pages where Chinese characters cannot be selected are scanned copies and should be OCRed first before editing, otherwise the text cannot be modified.
Content copied from English PDF often contains hard line breaks and extra spaces. Delete the line breaks at the end of the line and unify the spaces to restore the paragraphs.
When the entire page or table is misaligned and the source file is Word or PPT, you should go back to the original file and re-translate it to retain the layout.
Check terms, numbers, and dates against the original text before exporting. Machine-translated content requires manual review.

After translating a PDF paper or contract into Chinese, a common situation is that paragraphs are cut into pieces, table columns are misaligned, titles and text are squeezed together, or there are several extra blank pages. Many people subconsciously change tools and go through it again. The result is that the same problem often occurs again in another place.

This article is written for readers who need to deliver a translation but must retain the original format. The core idea of PDF post-translation format processing is: first determine which layer the problem belongs to, and then decide whether to change paragraphs and pages in the PDF editor, or return to the source file and translate it again. Most errors only involve local parts, and direct modification is usually more time-consuming than redoing the whole thing.

After reading this article, you can determine which errors can be fixed by editing, which ones must be redone, and know how to export the modifications to others for use.

The layout is messed up. Do you want to re-translate it? One sentence conclusion

Direct modification is usually faster: as long as the text content of the translation is basically correct and the problems are concentrated on line breaks, individual cells or page levels, you should modify it in the editor. Only when the source file recognition fails or the overall layout collapses, is it worthwhile to re-translate.

your situation	Recommended practices	reason
There are only a few paragraphs with incorrect line breaks and sentence breaks.	Change paragraph text directly in PDF editor	The text content is basically correct and the scope of changes is small.
There are several misalignments on the page and there are many blank pages.	Delete blank pages or reorder pages using page operations	The problem is at the page level and has nothing to do with text
The scanned translation cannot select Chinese characters and cannot be modified.	Do OCR recognition first, then edit	Without a text layer, direct editing is not possible
Full-page tables and images are offset as a whole	Go back to the original file and re-translate the document	The overall layout is offset and local repair costs are higher.
After the revision, you still need to continue to type and finalize the manuscript.	After editing, export to Word	Formatting in Word is more flexible than adjusting in PDF

First determine what kind of problem it is

Spending two minutes sorting before starting can avoid repeated revisions in the wrong direction. The following table lists common symptoms, causes, and how to confirm them.

symptoms	Common causes	How to confirm
One sentence is cut into several lines, with gaps at the end of the lines	Hard line breaks were retained when copying the original text	Click into the paragraph in the editor and check whether the line break position is consistent with the original line width.
English words stick together, or the spacing between Chinese and English is abnormal	Font missing characters or character mapping issues	Copy a few words into plain text and see if garbled characters appear or spaces disappear.
No text can be selected on the page	Scans, pages are just images	Use the mouse to drag a piece of text. If it is not selected, it will be an image page.
Table columns are misaligned and cell text overflows	The translation is longer than the original text and stretches the page.	Compare the original page and the translated page to find out which column or row the overflow is from
The title is mixed with the text	Title is recognized as part of the text	Check paragraph by paragraph to see if the title is in a different paragraph than the main text.
The number of pages is incorrect, a blank page appears or the page order is reversed	Error while paging or merging files	Turn to the last page to check the number of pages, and compare the page order with the original text page by page.

Check and repair item by item

Process in the following order, first solve the problems that have a large impact and are easy to confirm, and then deal with the details.

Check the page number and page order first. Open the translation and the original text side by side and compare them page by page. Use the page operation to delete the extra pages, and rearrange the pages in reverse order using the page operation.
Determine whether it is a scanned copy. Use the mouse to drag a section of text. If it is not selected, it means that the page only has images, and OCR recognition needs to be done first before the text can be modified. OCR results should also be checked, especially numbers, formulas, and proper nouns.
Fix line breaks. Click into the cut paragraph in the PDF editor, modify the text directly, delete the hard line breaks at the end of the line, and connect the entire paragraph into a complete paragraph. Multiple line breaks within a paragraph can be processed at once without having to adjust line by line.
Unify the spaces and punctuation between Chinese and English. Content copied from the original English text often has extra spaces, and spacing between Chinese, English, and numbers is also prone to inconsistencies. Check paragraph by paragraph "," before and after ".", as well as the spaces between English words.
Handle overflowing table text. For cells where the translation becomes longer, the wording can be shortened without changing the terminology. If the entire table is misaligned, it means that the layout cannot be partially repaired, and you should go back to the source file and re-translate the page or the entire document.
Check titles and levels. Change the titles that were mistakenly incorporated into the text into independent paragraphs to make the hierarchy consistent with the original text. When there are a large number of titles, it is safer to check them one by one than to replace them as a whole.
Check before exporting. Check that terms, numbers, and dates are consistent with the original text, and read through key passages. Machine-translated content requires manual review and cannot rely solely on the translation itself.

When does it have to be re-translated?

Source file is Word or PowerPoint. The PDF itself has lost some layout information. Translate directly using the original .docx or .pptx file, retain tables, pictures, and formulas, and choose bilingual comparison or translation-only output.
The scan quality is very poor. OCR recognition errors will be propagated to the translation when the page is skewed, blurred, or heavily shaded. A clear original should be obtained before OCR and translation.
The entire document layout has been shifted. If almost every page needs to be adjusted, the time required to patch often exceeds the cost of re-translating, in which case re-translation is more cost-effective.

How to avoid it happening again next time

Find the original editable file before translating. Prioritize using Word or PowerPoint source files. Do not convert repeatedly from PDF.
For scanned documents, first confirm the clarity of each page and process skewed or blurred pages before performing OCR.
Choose the appropriate translation style based on the document type, such as academic, legal, technical or business. After the terminology is unified, the number of changes that need to be made later will be significantly reduced.
Long paragraphs should retain the paragraph structure of the original text. Do not manually split it into multiple paragraphs for the convenience of typesetting. It will be difficult to merge after splitting.
Export the translation in a unified manner after it is finalized to avoid exporting multiple versions while making changes, which may cause file confusion.
A quick check with the number of pages, titles and tables in the original document before delivery can detect most structural problems in advance.

❓ Frequently Asked Questions (FAQ)

1. After PDF translation format processing, can I change the words directly on the translation?

Can. Use the PDF editor in the desktop client to open the translation, click on the paragraph to modify the text directly, and after modification, you can export it to Word for further typesetting. The client can be downloaded from download.omnidoc.top.

2. How to adjust the English format of pdf copy?

English copied from PDF often has hard line breaks and extra spaces. The straightforward approach is to open the paragraph in the editor, remove the line breaks at the end of the line and unify the spaces. If you want to paste it elsewhere, you can paste it into plain text first to clean up the formatting, and then reformat it.

3. How to edit pdf translation online?

After exporting to Word, you can continue editing in the UniDoc web editor (app.unidoc.top), which supports Word-compatible editing. The paragraph modification of the PDF itself still needs to be completed in the PDF editor, and it is more convenient to use the two together.

4. Why can’t I select the correct words after translation of the scanned PDF?

Because the scanned pages are images, there is no selectable text layer. First perform OCR recognition to generate text, and then modify it. After the identification is completed, check it paragraph by paragraph. Numbers and proper nouns are especially prone to errors.

📝 Summary

The key to format processing after PDF translation is to first determine whether the problem is line breaks, pages, or overall layout offset, and then choose partial editing or re-translation. Most errors can be fixed directly in the editor, and only if there is a problem with the source file or layout itself is it worth going back to the original file and redoing it. Don’t forget to check the key content against the original text before exporting.

Open OmniDoc and use the PDF editor to work with your document