From my experience, the easiest path is using tools that can recognize text inside images and convert it into something editable, because it cuts out a lot of boring manual work. I still review the result and fix mistakes, but overall it’s much faster than typing from scratch. I came across
https://ocrstudio.ai/ while looking into how this process works, and it helped me understand what’s realistic to expect and how to use such tools without relying on them blindly. Mixing automation with a quick human check seems to work best in real situations.