Jump to content

(Archived) Server workflow direction for text recogntion in notes

Recommended Posts

Is there a good diagram somewhere showing how a note captured via camera gets to our evernote clients, showing where text recognition is done? My main problem is that hand-written notes are not even remotely being recognized well for me, and I'm not sure if the note I have in my client is the final version of the recognition that will ever be done. I also don't know if there is a way to send it back to the server to try again. Also, it would be nice to be able to send poorly recognized notes to a central server that can use it to improve the handwriting recognition software.

Link to comment

Once the service has processed one of your images, there's no way to tell it to "try again", since it would give the same results every time until we change the version of the recognition software we're using.

From the Mac client, you can see whether your images have been processed by showing "More Info" for the note. This isn't in the Windows client UI yet, but we plan to introduce similar functionality as part of our UI rewrite.

Image processing rarely takes more than a few minutes, so if you've synchronized an hour after the image was first uploaded, you almost definitely have the recognition results for your notes.

We are looking for good ways to allow people to submit images that did not recognize very well, so we can use this to help train our algorithms.

Link to comment


This topic is now archived and is closed to further replies.

  • Create New...