How to Convert Image to Text in Microsoft Word

Images are everywhere – WhatsApp forwards, restaurant receipts, client emails, and your employee ID card. Images are the most common form of communication across channels.

But how do you convert these images into editable Word files?

In this article, we will learn how to convert an image to text using Microsoft Word.

While this method doesn’t use OCR, we will also convert an image into a Word file using OCR and understand the differences.

---

Need a smart solution for image to text, image to table, image to excel, Word OCR, PDF to Word, PDF to text?

Check out Nanonets’ pre-trained data extraction AI for bank statements, invoices, customer orders, purchase orders, receipts, passports, driver’s licenses & or PDFs!

---

Using Microsoft Word

First and foremost, Microsoft Word does not have a direct option to convert a picture to text. 

However, using this hack, you can convert an image to editable Word by converting it into a PDF first.

Converting a simple image to text in Word

Let’s have a look at the steps.

I am taking an image (Figure A) in PNG format to convert into an editable text Word document. You can also apply the same process for JPG to Word.

Step 1: Open a new document in Microsoft Word and Insert the image.

Steps to follow: 

Right-click on the image > Click Copy (Ctrl/Command + C) > Click Paste (Ctrl/Command + V) (Figure B)

OR 

Go to a New file on Word > Open/Insert > Select the image you want to convert

Copy the image in to a new Word file. (Figure C)

Step 2: Save the Word document as a PDF

Steps to follow: File > Save as > Rename the file > Choose the format as PDF > Choose the location

Save the Word document with the image as a PDF file. Remember to save by a name you can remember so as to locate it easily later.

I renamed the document SaveImageAsPDF and exported it to the desktop.

Word will process the image in a few seconds to a minute and save the document as a PDF in your chosen location. (Figure F)

I double-checked the saved PDF document on my desktop. (Figure G)

Step 3: Open the saved PDF file in a new Word document

Steps to follow:

Return to Word > Search for the previously saved PDF file by its name > Open\

Word will now detect the text in the PDF and show it on the Doc. (Figure H)

This is how you can convert an image into editable text using Microsoft Word. You can now tweak the text as per your preference.

Converting a complex image to text in Word

Now let’s try the same activity with a more complex image. (Figure I)

I followed the same steps as above. Here was the result:

As you can see that while Word detected some text, the accuracy is extremely low. There are also a few snips of the image as Word was unable to extract text directly, not to mention the formatting issues.

Using OCR

Now, let’s try using OCR on the same two images that we used above.

OCR on a simple image

I tried Nanonets‘ online Image to Text converter that uses an in-built OCR engine. (Figure 1).

I downloaded the text in a TXT file and opened it in a new Word file. And this was the result. (Figure 2)

Pretty neat and accurate, right?

OCR on a complex image

Now, time for the real-test.

I tried the OCR on the same receipt we used earlier. It has complex formatting with logos, some handwritten text, and different layouts.

The OCR captured all the text precisely within a few seconds. Once I downloaded the text, I opened it in a Word document; and this was the final output:

I was surprised to see that the OCR detected the handwritten text and was able to populate different line items in the Word file fairly accurately.

Microsoft Word vs OCR

As you saw above, while it was easy to convert a simple image to text using Word, the same process did not work well on a complex image.

Note that Microsoft Word doesn’t use OCR, so it often gives inaccurate results if you try to convert complex pictures or images into editable text documents. You also need to spend a significant amount of time to manually edit the formatting before you can use the text for further computation or analysis.

While you don’t need additional software or an internet connection to convert pictures on Word directly, you need a paid Microsoft subscription. The Read-only Microsoft version doesn’t allow you to use features like Save, Open, etc.

Unlike Microsoft Word, Nanonets OCR works with near 98%+ accuracy on both simple and complex images. It’s more reliable and processes PDFs in a few seconds or minutes.

How to convert complex images into editable Word files

As we saw above, converting an image to accurate editable text in Word is extremely difficult.

While many image-to-word or image-to-text converter tools are available online, they work best only on a few simple images. Most lack the powerful OCR and deep learning capabilities to convert hundreds of complex images. 

This is where AI-enabled OCR software comes to the rescue.

While a few simple images can be converted to editable text documents by using Word or online converter tools, the challenges arise in converting:

While Microsoft Word won’t be able to accurately you convert these images to text, Nanonets OCR uses deep learning capabilities to extract text from them.

All you have to do is:

1. Sign up on Nanonets.
2. Choose from either a zero training AI model or one of the pre-built AI models depending on the type of image.
3. Upload all your images and wait for Nanonets to extract text from them.
4. Open the converted image to verify the converted text and line items.
5. Make customizations as per your needs.
6. Export the data in your preferred format (Word, TXT, CSV, XML, XLSX)

Final word

We learned how to convert images into editable text on Word. While it is not a single-step process, converting images into text is much easier using Word than entering it manually using Copy + Paste. 

However, while converting images into editable text is possible in Word, the accuracy is low. The post-processing requires significant manual effort. Using OCR and Deep Learning, extracting text accurately from images is much more accurate, efficient, and reliable.