We have a client who is scanning large batches of documents (Laserfiche Cloud)
We would like to schedule the OCR processing for overnight
Any ideas?
Thank you,
We have a client who is scanning large batches of documents (Laserfiche Cloud)
We would like to schedule the OCR processing for overnight
Any ideas?
Thank you,
Laserfiche Cloud does not currently support scheduling OCR. All documents without text are queued up for OCR.
Why does OCR have to only occur at night?
Thank you, Miruna
They scan large batches and it is slow as it OCRs the content
Oh, you're OCRing it on the Laserfiche Scanning (or Quick Fields side). You can turn that off for Laserfiche Cloud. We have a background OCR that takes care of all new image documents coming in.
Hey Miruna,
Is this background OCR on by default, or is there a setting that has to be enabled to take advantage of this?
It's on by default.
Hi Miruna,
That's great that clients that are using Cloud don't need to worry about OCR on import. Can you elaborate on the schedule or how those documents are queued up to be OCR-ed? Is it something that will happen immediately or will it take a few hours to complete?
Thank you!
Molly
We have an OCR farm, all new documents with image pages, but no text, will be added to the queue. How fast it happens will depend on the load on the farm, it's usually done within a day or so, but large import batches can have an impact.
I would recommend OCRing upfront on the client side in the cases where the text is critical for filing or routing the document and only relying on the background OCR when text is just a "nice to have" thing for searching.
Edit: On-demand OCR is also available through Capture Profiles, so if you'd rather not OCR on the client side at import time, but need the text for routing, that's another option.
Does the volume of imports needing OCR from "Your Cloud" impact the rate at which the imports on "My Cloud" will be OCR'd?
Yes, at this time, the OCR farm is shared across all tenants. Its purpose is to generate text for all documents to make searching easier.
See comment above about capture profiles.
Will LF release any cloud updates anytime soon in order to batch OCR entries that are currently in the repo without OCRed text?
Also, what automated workarounds are currently being used by cloud customers for this issue, if any please?
Thanks in advance.