How to use the office file media extractor
- Choose a Word, PowerPoint, Excel or OpenDocument file: .docx, .pptx, .xlsx, .odt, .ods, .odp or .odg.
- Read the list of media files: the name, the type read from the first bytes and the size.
- Choose Download on a row to save one file, or Download all as ZIP to save every file at once.
- Use Download list (.csv) to keep a record of the names, paths and sizes.
Where the pictures live
A modern Office file is a ZIP package. Inside, the text and layout are XML files, and pictures, audio and video are stored in a media folder as separate files in their original form: word/media in a Word document, ppt/media in a presentation and xl/media in a workbook. OpenDocument files keep pictures in a Pictures folder. Because these are the original files, saving them gives you the full resolution image that was inserted, not a screen-sized copy.
This tool opens the package, reads each file under the media folder and hands it back byte for byte, with no recompression. The type shown comes from the first bytes of the file, not its name, so a PNG saved with a wrong extension is still labeled as PNG. A Word file with one PNG and one JPEG picture, for example, lists word/media/image1.png (78 bytes in the test file) and word/media/image2.jpg.
Safety and limits
Pictures are not drawn on this page, and SVG drawings are listed as text files and never opened, because they can contain scripts. Files that unpack to more than 400 MB, more than 5,000 entries, or entries that unpack larger than their header says are refused. Password-protected packages are skipped. Embedded objects such as other documents are not extracted. For the text of a Word file, use the DOCX text extractor. To open an ordinary ZIP archive, use the unzip extractor, and to pull pictures from a PDF, the extract images from PDF tool.
Assumptions and limits
- The file is a ZIP-based Office or OpenDocument file. Older .doc, .ppt and .xls files are not supported.
- Only files in the standard media or Pictures folders are listed. Thumbnails of OpenDocument files are skipped.
- Names that repeat are made unique with a number in the ZIP.
- Everything runs in your browser. Nothing is uploaded.
Frequently asked questions
Why can I not just rename .docx to .zip?
You can, but this tool also checks that the package is a real Office file, reads the real type of each media file and saves them in one step without a desktop archive program.
Are the extracted images the originals?
They are the files stored in the document. If the author inserted a large photo, you get that photo. If the program recompressed it when saving, you get the recompressed version, which is what the document holds.
Why is my file listed as having no media?
The document may contain only text, or its pictures may be linked from elsewhere instead of embedded. Linked pictures are not inside the package, so they cannot be extracted.
Can I extract charts?
Charts are stored as XML parts and not as media files, so they are not listed here. Copy them inside the program, or export the document to PDF.