Using AI to transcribe hand written notes

A digital representation of a handwritten note being transcribed into an editable note using OCR technology. The image shows a piece of paper with neat handwriting on one side, transitioning into a digital screen with typed text on the other side. The transition effect includes glowing lines and bits of data flowing from the paper to the screen. The background is sleek and modern, with subtle tech elements like circuit patterns and light beams to represent the technology at work.

An image depicting a handwritten note being transcribed using OCR and AI. Generated by DalE

One beneficial use case of AI is performing handwritten note transcriptions. I have used freely available AI resources to transcribe a list of donations written by my sister. My sister's handwriting isn't the best in terms of legibility, but it is legible enough with just a few letters that will take some time to decipher but are usually discernible for normal people to guess given contextual cues. I have used Google's Gemini, Opera's Aria, Microsoft's Copilot, and OpenAI's ChatGPT to do these tests.

The results were impressive and although the output will still need to be carefully proofread by a human being, this will greatly help in simplifying the workflow of people that have to deal with this sort of thing on a regular basis. You can also ask the AI to do follow-up tasks such as formatting the transcribed text into a tabular format.

Interestingly, of the 4 free AI assistants I have tested, Opera's Aria seems to produce the most accurate results for my particular test case.

I will definitely be exploring other things that AI can be used for my day-to-day tasks.