Cloud Development Kit

Document for content types

Hi all,

I'm evaluating MyScript CDK.
The textInput parameter for text recognition has contentTypes parameter.
Where can I find a document for each content type?
I can't find out the meanings of the content types grm, mul/gesture, mul/number, and mul/phone_number.

Thanks


Dear Kurima Yoshida San,

thank you for contacting us.

Indeed, using the /api/v3.0/recognition/rest/resource/language/{language}/contenttypes.json function allows to get the list of resources available for each language.

We do not provide with a specific description of each resource, as the name of the resource is normally implicit. You are of course right to ask more questions. To answer these:
-The grm resource is to allow to enter several times in a lexicon. Indeed, by default, a lexicon, only allows to recognize one input of the lexicon: Let's say you are using the first_name lexicon, and input Paul and George. Basically, in the lexicon, these are two entries, and as you are not able to have multiple inputs, you are not likely to get a proper result. Attaching the grm with the first_name lexicon will allow to enter several times in the lexicon, and you should be able to get Paul George. Please note some resources such as first_name.extended, last_name.extended, text and text.lite already include the grm.
-Regarding the mul/gesture resource, as gestures are not implemented in the MyScript Cloud, it is currently useless.
-The mul/number is to recognize numbers. Ideally, we would recommend you use the specific number resource for each country.
-And for the mul/phone_number, we also recommend you use the specific one of each country.

Let us know if you have further question.

Best regards,

Olivier

Dear Olivier,

Thank you for your reply.
I understand the use of the resources.

Let me ask another related question.

I want to disable recognition correction by lexicon.
Because I want accept users' spelling mistakes.
I want recognize handwriting as users wrote.

Can I set a empty array for contentTypes?
Or, is there any other way to disable lexicon?

Best regards,

Kurima

Dear Kurima Yoshida San,

In order to disable any lexicon, you could set the grm resource only.

Nevertheless, we warn you that our technology has not been designed to do character by character recognition, and recognition result is not likely to be accurate.

Best regards,

Olivier

Dear Olivier,

Thank you.
My program set the grm resource is working well.
And I also understand your warning.

Best regards,

Kurima

Dear Olivier,

Can i know the difference between grm and text? grm is behaving differently in character by character and word mode?

Is there any document available which explains the contenttype differences?



Dear Balachandran,

currently, the TEXT resource is our default resource, which we recommend you use to recognize "standard notes". It can recognize standard words of a given language, digits, first/last names, dates...

Regarding the GRM resource, it is already included in the TEXT resource.  This resource is to be used when attaching a single lexicon (i.e. list of words), and you want to be able to recognize several entries of your lexicon in one request. Here is an example:
-You have a lexicon containing words such as, paul, john, jean...
-If writing "paul" only, it will be recognized.
-If writing "paul john", without the grm resource, it will not be recognized, as with a lexicon, by default you can only recognize one entry at once.

Best regards,

Olivier
Login or Signup to post a comment