OCR documents that contain transliteration of Sanskri

[Leo Rivers]

Greetings

I am trying to OCR some documents that contain transliteration of Sanskrit, meaning that they are Latin characters with a set of about 10 diacritic marks above and below them, which are not standard.

In order to make my OCR software (ReadIris Pro) learn the right characters, I need advice.

First, have I the correct product to do what I need to do?
----
Readiris 12.0 (build f3) - i64
---
Model Name:    Mac Pro
  Model Identifier:    MacPro4,1
  Processor Name:    Quad-Core Intel Xeon
  Processor Speed:    2.66 GHz
System Version:    Mac OS X 10.6.8 (10K549)
  Kernel Version:    Darwin 10.8.0
  Boot Volume:    Macintosh HD
Readiris:

  Version:    12.0
  Last Modified:    6/28/10 8:01 AM
  Kind:    Universal
  64-Bit (Intel):    Yes
  Get Info String:    Readiris 12.0.5 (build 1ef) Copyright 1987-2009, I.R.I.S.

  Location:    /Applications/Readiris Pro 12/Readiris.app

SAMPLE TEXT :

"The following brief statement regarding Vasubandhu’s religious view is limited to information obtained from his Sūtra-commentaries translated into Chinese by Bodhiruai during the first half of the sixth century A.D. These include T.1519 (SPU), T.1522 (Daśabhūmika-sūtra-śāstra) & T.1524 (SU), T. 1525 (Gayāśirṣa-sūtra-tīkā), T.1526  (Ratnācuḍaparipṛcchā-sūtra-catardharma-upadeśa), T.1532
(Viśeṣacintiparipṛcchā-sūtra-upadeśa) and T.1533 (Dharma-cakrapravartana-sūtra-upadeśa)."

[kirtu]

Well, when you scan the document does it scan correctly?

Kirt

[Kaji]

Would installing the Sanserif Pali font (http://www.dharanipitaka.net/2011/2008/ ... nspali.ttf) help?

[viniketa]

Leo - I'm not familiar with Readiris, I typically use ABBYY Fine Reader (and I use a PC). However, you likely need a 'language pack' for Readiris that includes all Extended Latin characters. English language does not. French, Spanish, German do.

Hope this helps.