Hello all,
Please, can you advise which Abbyy version is Language Weaver currently using? Abbyy has recently introduced some AI-enhancements, so I was wondering whether Language Weaver is using that version of Abbyy.
Also, would it be possible to create a source OCR in Language Weaver, too? If the option is already there and I have somehow overlooked it, could you please point me to the right settings?
Thank you :)
Veronika
Hi Veronika Zbrankova ,
In Language Weaver, we're integrating the latest version of the ABBYY Finereader SDK (12.5).
By default, we include a standard PDF converter which is providing a solid conversion for most of PDF sources, but ABBYY is providing optimized and improved PDF conversion for scanned documents and non Latin script languages. When you purchase the ABBYY option, you can let our SmartSelection decide what's the best library to use (rules are described here: https://docs.rws.com/791593/1090329/language-weaver/selecting-pdf-conversion-options).
It is not possible to define a different OCR source. If you would like to use a different PDF converter, you would need to perform your conversion (usually to the docx format) first, and then process the converted output with Language Weaver.
Thank you.
Hi Arnaud Simon ,
thank you so much for your answer, it was really helpful :)
As for the second question about OCR source, I'm not sure if I phrased it correctly, so I'll try again. I didn't mean to use a different OCR source, but rather use Language Weaver solely for OCR-ing a file - e.g. when I drag'n'drop a PDF file in Language Weaver, it will translate the PDF and I can download a Word target file, already translated. I was wondering whether it is possible to also download just the OCR-ed source file, though. Sometimes, the client is interested not only in translation and outputs, but also in the source itself, usually when they only have a scanned PDF document.
Thank you again for your help
Hi Arnaud Simon ,
thank you so much for your answer, it was really helpful :)
As for the second question about OCR source, I'm not sure if I phrased it correctly, so I'll try again. I didn't mean to use a different OCR source, but rather use Language Weaver solely for OCR-ing a file - e.g. when I drag'n'drop a PDF file in Language Weaver, it will translate the PDF and I can download a Word target file, already translated. I was wondering whether it is possible to also download just the OCR-ed source file, though. Sometimes, the client is interested not only in translation and outputs, but also in the source itself, usually when they only have a scanned PDF document.
Thank you again for your help
Hi Veronika Zbrankova ,
With the embedded PDF Conversion libraries (ABBYY or the standard one), it is not possible to provide the OCR version of a document only: it has to be translated, as part of the overall processing workflow. This is not a technical limitation, but a commercial one. We can use OCR as part of the translation workflow, but we can't offer it as a standalone service.
Thank you
Thank you for your answer, Arnaud, I understand :)
Translate
