FAQ
Common questions and answers for all PDF tools
FAQ now works as an answer page as well as a navigation hub, surfacing selected questions and practical guidance before users jump into a tool.
什么时候先看 FAQ，而不是直接进工具页
如果你已经知道自己要做的是 PDF 转 Word、压缩、OCR、合并或拆分，直接进工具页通常最快。 但如果你还不确定该选哪种工具、扫描件要不要先 OCR、压缩和拆分谁先做、或为什么转换结果和原文档差异很大， 先看这里会更省时间。
这页不是单纯的链接目录，而是一个帮助中心入口：每个工具先给出代表性问题与答案， 让你在点击前就知道适用场景、限制条件和常见误区，减少试错成本。
如果你的目标是判断“现在该转 Word，还是先 OCR，还是先拆分再压缩”，FAQ 页比直接点进工具更适合做第一轮判断。 它把不同任务的前置条件、适合的输入类型、常见失败原因和下一步建议集中放在一个入口里，减少因为误选工具带来的返工。
对经常处理合同、扫描件、报告、课件和表格类 PDF 的用户来说，这里也可以作为一个快速分流页： 先搞清当前问题属于编辑、识别、压缩、拆页还是导出图片，再进入对应工具，通常比逐页试错更快。
PDF to Word
8 questionsThe Word file looks quite different from the PDF — why?
PDF and Word use different layout engines. Heavy graphic layouts can shift; for best results stick to text-based PDFs and double-check the headers, footers and TOC after conversion.
What slows the conversion down?
Page count, the number of images and the file size all matter. Documents under 10 pages usually convert in well under 30 seconds.
Can I upload password-protected PDFs?
Not yet — please remove the password first using Acrobat, Preview or a similar tool.
PDF to PPT
8 questionsWill the slide count change?
By default each PDF page becomes one slide; we do not auto-split or auto-merge slides.
Is the text really editable?
Yes for text-based PDFs — text becomes editable text boxes. Scanned PDFs keep words as images.
How is the slide aspect ratio decided?
We follow the source PDF page ratio, typically 16:9 or 4:3.
PDF to Excel
8 questionsWill numbers be treated as text?
We try to type them as numbers, dates or percentages. If something arrives as text, use Excel's Text-to-Columns to fix it.
What happens to a table that spans multiple pages?
We try to align by header and merge them; if the header changes between pages, you may get separate tables.
My PDF has several different tables — what mode should I use?
Use "split per page" so each table sits in its own sheet; "single sheet" works best when the layouts repeat.
PDF OCR
8 questionsIt still looks like a scan after OCR — is that right?
Yes. We add an invisible text layer over the page image, so it looks identical but is now searchable and copyable.
Roughly how accurate is recognition?
Clean printed scans are usually much more accurate than blurry, skewed, or handwritten pages.
How do I get an editable document?
Run OCR first so the PDF is searchable, then run the file through PDF → Word for a real editable copy.
PDF to Markdown
3 questionsCan scanned PDFs be converted to text?
This tool does not run OCR. If the output is sparse, use PDF OCR first and convert again.
Why does external image mode download a ZIP?
Markdown uses relative image paths, so the .md file and image folder must travel together.
Are tables preserved?
The converter outputs GitHub Flavored Markdown tables where possible. Complex layouts may need manual cleanup.
PDF Compress
8 questionsWill text look blurry afterwards?
Default compression only down-samples images; text remains in its original vector form and looks identical.
Why didn't the file shrink much?
It may already be compressed, or be vector-and-text dominant with no raster bulk to drop.
Can I pick a compression level?
We currently apply a single near-lossless preset; tunable levels are on the roadmap.
PDF Merge
8 questionsHow big and how many pages will the merged file be?
Roughly the sum of all source pages and sizes; we don't compress during merging.
How many source files can I merge at once?
There's no fixed cap, but a few dozen at a time is the comfortable upper bound for speed and reviewability.
I picked the wrong order — can I fix it?
Drag rows freely until you click "Merge PDFs". Reorder as many times as you like.
PDF Split
8 questionsHow do I list multiple non-contiguous ranges?
In "by page ranges" mode separate ranges with commas, e.g. 1-3, 5, 8-12 produces three subset PDFs.
How are output files named?
By page span by default, e.g. part_1-3.pdf; in equal-split mode they are part_1.pdf, part_2.pdf, etc.
Can ranges overlap?
Yes — the overlapping pages will appear in each subset.
Image Export
8 questionsWhat's the difference between "extract embedded" and "render each page"?
The first only returns images that were embedded as objects in the PDF, at their original resolution. The second renders the whole page (text included) as a PNG.
Are the exported images watermarked?
No — exports are clean, with no service watermark.
Can I keep vector graphics as vectors?
Vector export to SVG/EPS isn't supported yet; vector content arrives as bitmap.
Watermark
8 questionsWill every page be watermarked?
Yes by default. Enter a range like 1-3,5,7- to watermark only selected pages.
Can I use Chinese or Japanese watermark text?
Yes. The backend automatically tries a CJK-capable font when non-ASCII text is detected.
Which image formats are supported?
PNG, JPG, and JPEG are supported up to 5MB. Transparent PNG areas are preserved where possible.
Signature
2 questionsIs this a legal digital signature?
No. It only embeds a visible image and does not provide certificates, timestamps, or tamper-proof validation.
Are signatures stored?
Server-side files are temporary and cleaned after about one hour. The browser signature library stays only on your device.
Convert to EPUB
3 questionsCan scanned PDFs become EPUB text?
Not directly. If the extracted content is sparse, run PDF OCR first and convert again.
Which ebook formats are supported?
This release supports EPUB only.
What if I do not upload a cover?
The first PDF page is rendered as the cover; if that fails, a simple text cover is generated.