Slow conversion of scanned pages to searchable pdf

[S M]

Hi,

As the title says, I was wondering if there's anyway to speed up the conversion process.

I'm moving to you from PaperPort 12 because a) their support is abysmal and b)their program keeps losing any scans >20 pages.

I am scanning 99% of the time black and white pages of text.

My current settings are:
Scanner KODAK i2600
TWAIN
Paper source: Feeder
Size: Letter (8.5 x 11in)
Resolution 300dpi
bit depth: black and white
scale 1:1
No advanced settings besides default.

It takes something like 15 seconds per page - when I'm scaning something like 80-120 page documents this means processing takes a long time. Certainly far longer than PaperPort managed it in.

So, do you have any advice or tips for speeding this process up?

All the best and thank you for your software,

Seb

[Ben Olden-Cooligan]

There's no easy way to speed it up right now. You could try reducing the resolution, but of course that's not ideal.

However, I am currently working on making OCR faster. I can't provide an ETA but if I can get it done it should be several times faster, depending on your computer.

[Rodrigo Pozzebon]

It is already possible to provide an ETA?
this change is very expected

[Werner]

I'm in the same situation... Paperport is an old, buggy and expensive software and I'm looking for a replacement. NAPS2 is partely a good replacement and I hope, that there will come some functions for document management in it. An easy improvent would be if NAPS2 remembers the file name of the imported PDF and give this as the default name when saving. Often I open/import an existing PDF, scan one or more pages and save it again with the same name.
It's very slow, if I have many pages in this PDF. Is it not possible to do the OCR only for the new scanned pages and save the 'old' pages without changes?
Another wish: it would be nice if I can import PDFs, which are not created with NAPS2...
@Ben:Thank you for this nice software.
...
A day later another (silly?) idea: what's about a function for saving with the option to append the actual scans to the begin or end of an existing file? In this case there would be only the need for recognizing the new scanned sites and the saving would be much faster?!?

And the last wish (for today): bookmarks would be a nice feature, too. If I could enter a short text in the context menu of a scanned page as bookmark, it would be possible to identify the pages faster.

[Ben Olden-Cooligan]

Importing PDFs not created with NAPS2 isn't going to happen for technical reasons, but I'll consider the rest for a future version.

[Ben Olden-Cooligan]

I still need to do some more testing before publishing the next version, but if you want you can try the faster OCR with this test version.

[Rodrigo Pozzebon]

Very good
For comparison (15 pages - OCR Portuguese)
NAPS current: 41 seconds
Kodak SmartTouch for Kodak i2600: 14 seconds
NAPS beta: 17 seconds

Excellent job
Thank you

[Ben Olden-Cooligan]

The faster OCR has been published in NAPS2 5.3.0.

[Werner]

@Ben: thank you very much, it's much faster... Even I've saved a PDF with 81 pages/200dpi in less than one minute. Before it needs much longer. So I've now more time for my suggestions ;-)

• whats about placing a cursor at the point, where the next scanned page should be inserted? Similar to the cursor which appears while moving pages... If no cursor is placed, NAPS2 could use the default behaviour (append page at the end). This cursor could also be used for importing pages from file at the desired position.