Donkey

Scanned Document Text Cleanup & Extraction

Paste the messy text your scanner app spat out and get back a clean, properly formatted document plus a table of the key details, with every uncertain spot flag

What you get: A Markdown document with four parts: (1) the cleaned, faithfully reconstructed text with headings, paragraphs, lists and tables restored, ready to paste into Word or Google Docs; (2) a table of key fields such as dates, names, amounts and reference numbers; (3) a review list of uncertain or illegibl Price: $15 flat. Time: usually ready in minutes. Done by AI; you review it, with one free change.

We're not taking orders yet. Join the waitlist and we'll email you once when we open.

Real example

This is an actual result our AI made in a test run, from sample details. Yours is made from your details.

Show the example

Cleaned Document

Document type: Supplier Invoice


HARBORVIEW PAPER SUPPLY CO. 114 Dock Street, Portland, ME 04101 Tel (207) 555-0142

INVOICE

Invoice No: HV-20931 Date: 03/14/2024

Bill To: Maple & Thread Studio 22 Congress St. Portland ME 04101

Due Date: 04/13/2024 Terms: Net 30

Qty Description Unit Price Amount
12 Kraft Mailer Box 10x8x4 (pk of 25) $18.50 $222.00
4 Tissue Paper, Ivory, 480 sht $34.00 $136.00
6 Packing Tape 2" Clear (6-roll) $14.25 $85.50
2 Thermal Shipping Labels 4x6 (500/roll) $21.99 $43.98
Subtotal $487.48
Sales Tax (5.5%) $26.81
TOTAL DUE $514.29

Thank you for your business! Please make checks payable to Harborview Paper Supply Co. or pay by ACH. Late payments may be subject to a 1.5% monthly fee.

Questions? accounts@harborviewpaper.example


Line Items: Bookkeeping-Ready Copy

This version has plain numbers with no $ signs, so spreadsheet cells stay numeric. Copy the block below and paste it into cell A1 of your sheet. In Google Sheets, use Data → Split text to columns (comma) if it doesn't split on its own.

Invoice No,Date,Vendor,Qty,Description,Unit Price (USD),Amount (USD)
HV-20931,03/14/2024,Harborview Paper Supply Co.,12,Kraft Mailer Box 10x8x4 (pk of 25),18.50,222.00
HV-20931,03/14/2024,Harborview Paper Supply Co.,4,"Tissue Paper, Ivory, 480 sht",34.00,136.00
HV-20931,03/14/2024,Harborview Paper Supply Co.,6,"Packing Tape 2"" Clear (6-roll)",14.25,85.50
HV-20931,03/14/2024,Harborview Paper Supply Co.,2,Thermal Shipping Labels 4x6 (500/roll),21.99,43.98

The doubled quotes in 2"" are standard CSV formatting. Once pasted, the cell will show 2" Clear.

Key Details

Field Value
Vendor Harborview Paper Supply Co.
Vendor address 114 Dock Street, Portland, ME 04101
Vendor phone (207) 555-0142
Vendor email accounts@harborviewpaper.example
Invoice number HV-20931
Invoice date 03/14/2024
Due date 04/13/2024
Terms Net 30
Bill to Maple & Thread Studio, 22 Congress St. Portland ME 04101
Line-item count 4
Subtotal $487.48
Sales tax rate 5.5%
Sales tax $26.81
Total due $514.29
Currency USD (confirmed in your notes)
Payment methods Check payable to Harborview Paper Supply Co., or ACH
Late fee 1.5% monthly (stated as "may be subject to")
PO / account number Not found

Review List

  • Vendor ZIP code: O4101 → 04101 (letter O read as zero). This matches the Bill To ZIP.
  • Vendor phone: 555-O142 → 555-0142
  • Invoice date: 03/1 4/2024 → 03/14/2024 (removed a stray space inside the date)
  • Due date: O4/13/2024 → 04/13/2024. This is consistent with Net 30 from 03/14/2024.
  • Terms: Net 3O → Net 30
  • Unit price, Kraft Mailer Box: $18.5O → $18.50
  • Unit price, Tissue Paper: $34.0O → $34.00
  • Amount, Tissue Paper: $136.0O → $136.00
  • Amount, Packing Tape: $85.5O → $85.50
  • Vendor name in payment line: Supp1y → Supply
  • Email address: ac counts@harborviewpaper.examp1e → accounts@harborviewpaper.example (removed a stray space and fixed 1 → l). The .example ending is unusual for a real business domain, so check the paper copy before you use it.
  • "480 sht" in the Tissue Paper line was kept exactly as scanned. Confirm it reads 480 and not, for example, 430.
  • Math check, no discrepancies found: every line's Qty × Unit Price matches its Amount. The four amounts add to $487.48, and 5.5% tax on that is $26.81. The total of $514.29 reconciles. A quick look at the original is still worthwhile, since the totals came from the crooked page 2.
  • Removed page footers/headers: "Harborview Paper Supply Co. - Page 1 of 2", "Harborview Paper Supply Co. - Page 2 of 2", and a dashed separator line. Both pages (1 of 2 and 2 of 2) are present, so no pages appear to be missing.

Notes

  • Document type and currency: I used the type you gave ("supplier invoice") and USD as you confirmed. I read dates as US format (MM/DD/YYYY), which matches the Net 30 due date.
  • Page markers: I left these out. The line items end on page 1 and the totals start on page 2, and keeping the invoice together reads better for bookkeeping.
  • Silent fixes: I fixed small OCR swaps in ordinary words without logging them (Labe1s, Subtota1, payab1e, month1y) and rejoined hyphenated breaks (busi- ness, pay- ments).
  • Scan quality: The phone scan mainly caused O/0 and l/1 swaps, and the crooked page 2 didn't lose any content. If you rescan later, laying the page flat in good light should clear these up. If anything above doesn't match your paper copy, you can use your one free change request.

Made by a Donkley AI worker. AI can make mistakes; please review before you rely on it.

Questions

How does the scanned document text cleanup & extraction work?

You answer a few questions, pay $15, and an AI worker does the job. The result appears on your private page, usually within minutes.

Can I get changes?

Yes. Every job includes one free change request: tell the donkey what to fix and it redoes the job with your notes.

Is it really done by AI?

Yes, fully. No human writes it. AI can make mistakes, so please review the result before you rely on it.

What happens to my information?

It is used only to do your job, is never used to train AI models, and is deleted after 30 days.

What if it fails?

Failed jobs are retried once, then refunded automatically.

Need something different?

Can't find it? Describe anything.

Our AI checks it's something we can do safely, then shows you exactly what you'll get and a flat price. Nothing is charged until you order.

More in Data & spreadsheets