I am currently trying to output all text from an image. I made this work using tesseract.js but now i faced a problem. I dont know how to analyze a conversation photo. If I just put it thru tesseract it outputs all the text but I need to somehow split it into the sender and reciver
For example with the default tesseract function it will output enter image description here
I dont think I can make it to the party? Are you sure?
But I need it something like
reciver: I dont think I can make it to the party Sender: Are you sure?
Or something like this so I can distinguish them somehow
Any idea on how I could do it? It would help me a lot. Thanks in advance
The only thing I thought I could "split" the image in half and read everythin on the left as reciver and everything on the right as sender but I have this problem where in some convos some text is bigger and passes the half of the screen