Watch the user experience of ABBYY FlexiCapture Standalone
Hello. Today I'm going to show you the user experience with ABBYY FlexiCapture, the Standalone version. It's a very simple tool. I think you're going to like it. It's going to flow very easily, so let's get started.
Now, the first thing we're going to do is, we're going to open up, what we call, the operator station and you can see that we have two different stations. One is the administrator station. That's where we do all of the behind-the-scene setup for a project. But the actual user experience happens, in what we call, the operator station. I'm going to hit the operator station and we're going to load the project. We can either browse or just open the project here. I'm going to go ahead and just open. The first thing we need to do is, we need to just follow this toolbar across. So, each one of these buttons, kind of, indicate the flow of how a batch, or a group of documents, would flow through the software. We're going to load some images, and literally, we select either one or multiple images. This is actually the image that we're going to take. This is a PDF document that contains multiple versions of this form for us. We're going to just double click it and the software's going to create a batch automatically. You can see here it put a new batch and what it's going to do is add that file to the batch.
It's going to add that and then the next thing it will do is, it will actually automatically do this recognize phase for us. When it recognizes a document, that's when the software extracts the details. It tries to recognize what kind of document type it is and then extract those details for us. That's what it's in the process of doing right now. Okay? Now, once the software is done recognizing it, it's ready for verification. Now, a lot of options from here on out are customizable, based on what you want to experience within the software. I'm going to show you a couple of different ways to run verification.
The first way is, we can simply just hit this button and this is a, what we call, a group verification process where we just go group by group, or field by field, and the software pulls all of these fields from the documents and asks us to determine if all of these, for in this case, are check marks. All of these look like we could consider them valid check marks and so, what we'll do is, we'll confirm them. In this case, it wants us to confirm when check marks aren't checked. In this case, we may want to say, hey, this one actually isn't considered a check mark, so that the software will consider that one checked now when we process it. This is what we call group verification. You can see it, it's telling you what the software recognized as the character and here are the different images within the different files that it recognized as well, and it's just asking us to determine which one of those matches.
We can literally go field by field and we can see here the percentage that we're at based on the batch and document that are here as well. For example, the software recognized these three characters as two, while we can assume that these are ones, so we can actually mark those as suspect characters and continue moving on, or we can even correct them and then we can tell the software these are ones, not twos, and therefore, it will save that for us. This is group verification and we just literally walk our way through until we're 100% verified. It's a very common and fast way to get through verification.
The other way, if I close out of this, is to just double click the batch. When I double click the batch, you can see that one file contained four separate documents that we extracted. Literally, what I can do is, just walk one by one, double click them and from a user experience, we see what the software extracted on the left, versus what the document looks like on the right, and we compare. This is what we call, what the software extracted is correct. So, we would walk through every single character. Any suspect characters are in red or anything underlined in red does need some sort of attention or is there for us to get our attention, just so we, as humans, realize that there is either something that requires us to look at or needs our attention.
We would literally just continue walking through the software. Now, the software also has built-in rules within it. These are rules that are customizable. They're up to you, but in this case, the software is saying, the first names don't compare on a couple different of the pages of this document and it really wants that to happen. All we need to do is, we can see if we just click on the field, it tells us where these fields are at, that must match. In this case, first name is on two or three different pages. Apparently, three pages here and it wants us to compare that they all look like the exact same name. Those look good. Here's the one that the software does not match in the rule, so all we need to do is override that ourselves, and when we do that, you can see that error goes away. You can save your changes as you move forward and then you would literally just continue walking through your batch, one by one, if you did not want to use the group verification.
Now, the other thing I will highlight is this confidence level concept. This confidence level is a character grade that we give for every character that the software extracts, it has its own confidence. What it can do is, based on its own confidence, if it is not confident that it extracted the right character, it will highlight the character in red and here's a couple of examples where the software, maybe, wasn't 100% confident, or confident enough, and therefore, it wants us, as humans, to verify that the software is correct. If for some reason it's not correct, in this case, this should say ave, we just simply override that field and when we override it, those results are what will get exported, not the incorrect results. It's just another step here of verification.
When we are done with verification, we just need to go to the next step up here, which is export. What we would do is, we'd highlight the documents that need exported and we would export. It's going to ask me if we want to save changes and then it's going to export the documents on which it can export. It does have an error and it's really just because of the error on this last one wasn't exported because I didn't fix all of the results of it. So, you can see here, we have rules here and I have the software configured that says, anytime we have a rule violation, make sure we don't export that, and really, the reason why we do that, is so that we don't get bad data when we're analyzing this downstream.
But, now that we've exported the other ones, so we can kind of go see where those are at, and really, all of this is set up in the administrator station on where documents are exported to, so not within the operator station, so we're going to assume that you know how to do that already. So, we will go look at the exported results here. You can see here, in this case, we have situations where we show every image, in other words, every separate document, and then we have all of the data as well. That's what these Excel spreadsheets are. Then lastly, if we redact some information on documents, we have those redacted details here as well.
That is the user experience within ABBYY FlexiCapture Standalone and it's a really, fluid, simple process, a very powerful process in the background as well. It gives you the ability to process with flexibility, your documents with your rules that are for your business. If you have any questions or if you've enjoyed watching this video, please let us know. You can reach us on our website at www.ufcinc.com. Thanks.