Recognition Server FAQs

This FAQ on ABBYY Recognition Server highlights some of the inquiries our customers most often pose. For first-hand experiences and case studies pertaining to Recognition Server, please visit the corresponding OCR software reviews page.

What is “OCR conversion”?

It is the process of using OCR software to convert image- or paper-based text into digital text. This is usually performed on large volume of text, typically with an end goal of searchable archives, as in these examples of medical data recognition and library archive publication.

Which languages can Recognition Server process?

Every major one…and a whole lot more. For basic text recognition, the official languages list comprises 43 with dictionary support and 133 without. That includes five East Asian languages, as well as a handful of artificial and programming languages, and even some chemical formulas. This case study is a great example of the power and breadth of Recognition Server’s multilingual capabilities.

For organizations like libraries and national archives, whose OCR conversion needs include historical documents, then the ability to process Gothic and other archaic scripts (in half a dozen European languages) is particularly valuable.

I’ve got lots of documents, but I simply want to turn them into PDFs, not to do anything fancy. Is Recognition Server overkill?

It’s true that Recognition Server has a lot of features. However, even when there’s no business need to automate batch separation, load balancing, bar code recognition, or output routing, users still enjoy what they find to be the most accurate OCR scanning software available, period. Even when the only criteria are accuracy, speed, and cost-effectiveness, a lot of organizations still choose ABBYY. As one wholesale conversion bureau put it, “[Recognition Server] is the gold standard for OCR accuracy.”

Furthermore, even if a client’s documents are simple and output/workflow requirements are minimal, processing capacity is a critical factor. The larger the input volume becomes, the more users will appreciate the ability to configure distributed scanning, remote and mobile access, and other features unique to enterprise-oriented solutions like Recognition Server.

Maybe PDFs won’t always suffice, after all. Which other formats can Recognition Server deliver?

Besides PDF and PDF/A, Recognition Server can also render documents in Office, XML-based, and common e-book formats, among others. Whatever the format, most output files can also be redacted and/or protected to ensure that the right information reaches the right eyes.

How do I know whether Recognition Server or FlexiCapture is more appropriate?

That’s a great question, especially since the underlying OCR technology is quite similar. It’s a little tricky to answer without knowing a individual client’s needs and reviewing examples of their documents. However, whether one buys FlexiCapture vs. Recognition Server is usually a matter of the end goal:

  • Recognition Server is tailor-made for creating searchable archives
  • FlexiCapture is oriented toward extracting text for use in business processes

The distinction can be subtle at times, and many organizations require a bit of each. And that’s exactly why our team is always here to chat. If you’re not positive which OCR software is most conducive to your needs, then a quick phone call will help resolve any uncertainty.

From which sources can I input documents?

Naturally, input can come directly from the Scanning Station(s) or any TWAIN- / WIA- / ISIS-compatible device. Additionally, the server can automatically monitor SharePoint libraries, “hot” folders, and mailboxes.

We use SharePoint a lot, but what value, exactly, does Recognition Server add?

Since SharePoint is widely used in just about every type of department, ABBYY has developed a very robust native integration. Most folks find that the biggest value is searchability of previously un-searchable documents (including ones in foreign languages!). Furthermore, if you have particularly high volumes of documents to input, then features like batch processing, scheduling, load balancing, and workflow management can further streamline things.

What can we do with the electronic, converted files?

Quite a few options are available, such as:

  • Choosing from a range of output formats (including PDF, XML, plain text, and Office file types, among others)
  • Publishing to SharePoint and/or network destinations
  • Sending via email

These choices can often be automated based on document metadata, which further reduces the need for manual involvement in the recognition process.

We’ve already got Google Search Appliance, so will Recognition server be able to work with it?

Yes, it will. This datasheet covers more details, but in brief, an OCR conversion server will a) make your image files immediately searchable by GSA and b) organize document metadata and index info into far more search-friendly formats.

Is this software useful outside of a business context?

It proves immensely valuable in a range of settings. Perhaps most salient among non-business customers are organizations like libraries and educational institutions, which often face pressure to digitize their collections, yet have little money with which to do so. Total page count licenses are available for one-time archival projects, although many institutions choose to make OCR conversion a ongoing part of their information management strategy. This brochure on Recognition Server in education may provide further clarification and inspiration.