Browse categories
Explore
Fiverr Pro
English
$
USD
I will convert chinese scanned PDF or image tables into clean excel data
PDF to Text Document Conversion Searchable Offline Libraries
I turn document collections into searchable, offline libraries. I cleaned, de-duplicated and indexed a 5,080-article / 9.57-million-character corpus into a single offline HTML file that opens with a d...
About this Gig
Numbers get trapped in PDFs, scans and screenshots. You need them in a spreadsheet you can actually filter and add up.
I build that file for you - and then I check it.
WHAT YOU GET
- Excel (.xlsx) or CSV, structured and formatted, ready to filter and sort
- Every value checked against your source file, not a raw OCR dump
- A short note listing anything unreadable or ambiguous in the original
- Free correction if any value does not match the source
WHAT I CAN HANDLE
- Chinese and English documents
- Text-based PDFs, scanned documents, screenshots and photos
- Multi-page tables, merged cells, tables that break across pages
- Messy sources: duplicated rows, broken headers, mixed formats
- Large sets, delivered in batches if needed
WHY ME
I run an automated extraction pipeline, then verify the output row by row. Reference project: 5,080 articles / 9.6 million characters extracted, cleaned, de-duplicated and indexed - with two source defects found and fixed during QA. Automation gives speed. Verification gives accuracy.
HOW I WORK
Send one sample page first. I will confirm the exact scope, price and delivery time before you order, so there are no surprises later.
Tool:
Excel
•
PDF editor
•
Other
My Portfolio
FAQ
Can you read Chinese documents?
Yes - simplified and traditional, printed or scanned. Chinese tables are my main focus. I also handle English and mixed-language files.
What if some values are unreadable in the source?
I will not guess. Those cells are flagged and listed in a note delivered with the file, so you always know what to double-check.
Will the output be editable?
Yes. You get a real .xlsx, .csv or Google Sheet with proper columns and headers - not an image, not a locked file.
Do you use OCR software only?
No. An automated pipeline does the heavy lifting, then every row is checked against the original. Raw OCR output alone is not delivered.
Can you handle handwriting?
Neat handwriting on clean scans, usually yes. Cursive or low-quality handwriting is hit and miss - send a sample page and I will tell you.
What counts as a page or a row?
One page is one side of one document. Rows are counted after extraction. Whichever number is hit first decides the tier.
