OCR file

Location on ribbon: Tab DVD Addin → Group File and Print → Menu PDF / OCR

Command ID: btnOCR

License required

This command needs an active DVDAddin account (Standard or PRO). During the trial period it is fully available.

What it does

The command sends an image or PDF file to a vision AI model (Gemini), reads every piece of text in it, and writes the result into the worksheet starting at the active cell — one text line per Excel row. It is the fastest way to get a scanned specification page, a photographed material certificate, or a PDF standard extract into Excel without retyping.

The prompt is written specifically for transcription: the model is told to keep the original language (Vietnamese stays Vietnamese, diacritics included), to rejoin lines that were broken only by column width, and explicitly not to summarise, interpret, or answer any question it finds in the document.

How to use

1. Select the target cell. This is the top of the result block — text flows down the same column from there.
2. Go to tab DVD Addin → group File and Print → menu PDF / OCR → click OCR file.
3. If no API key is configured, the add-in prompts you and stops. Set the key, then run the command again.
4. A file picker opens, filtered to .png, .jpg, .jpeg, .webp and .pdf.
5. Pick the file and click Open. The Excel status bar shows that OCR is running.
6. The result is written in a single pass into the range starting at the active cell, and the full text is also placed on the clipboard so you can paste it into Word or anywhere else.

Example

A supplier sends a scanned mill certificate (CO/CQ) as a PDF and you need its figures in your material tracking sheet.

Select cell B4 on the MaterialLog sheet, run OCR file, and choose CQ_Rebar_D16_Heat_A245.pdf. A few seconds later B4 downwards holds one certificate line each: mill name, heat number, grade CB400-V, nominal diameter, yield strength, tensile strength, elongation, and so on. You only have to move the rows you need into the right columns instead of typing everything again.

Notes

• The result overwrites the cells below the active cell. Make sure that area is empty, and note that the write typically cannot be undone with Ctrl+Z.
• A valid API key and an internet connection are required. Without a key the command stops with a prompt rather than pasting an error string into a cell.
• If the model returns an error, that error text is written straight into the active cell so you can see the reason; if no text is recognised the cell gets (OCR returned no text).
• This command extracts running text, not row/column structure. For tabular material use Extract Table instead.
• Accuracy depends on the source image: skewed, blurry, or heavily compressed scans produce spelling errors, so always double-check critical numbers.

Related

• Extract Table — same AI engine but preserves rows and columns, for tabular content.
• OCR Screen Clip — OCR whatever is on screen right now, with no file involved.
• Read PDF — open and work with PDF files directly inside Excel.
• Preferences — where the add-in's general options are configured.