How to correctly display characters in ePub for non-standard languages?

Nitnesh · 05-26-2017, 01:20 PM

Hi all,

I have a PDF as the source file for an ePub in a Ghanaian language and the text that I extract for the ePub does not display the characters/glyphs correctly (see attached screenshots).

Embedding the fonts will only help to display the characters in the ePub once I actually have the text, but how can I fix the source text in the xhtml?

Also in terms of which fonts to embed (if any?), Times seems to work in other ePubs that I have checked for this language, but I would appreciate any advice that you have with regards to fonts. Google Noto is supposed to work for Ghanaian languages as well, but do any of you have any (similar) experience with font problems and how did you fix them?

Thank you in advance!

Toxaris · 05-26-2017, 04:37 PM

Since the text files in an ePUB must be UTF-8, the Ghanaia glyphs are probably in there. Just use the correct characters and embed a font that includes these glyphs. You could try Charis Sil, it is rather complete as far as I know.
It should not be a problem at all.

Nitnesh · 05-27-2017, 06:55 AM

Hi Toxaris,

Thank you for your reply.

You are suggesting to key in the characters in the xhtml, right? This is what I have been doing so far, but it is a very long and strenuous manual job for long books, and prone to errors, as I do not speak the language myself.

Also, as I assume that this happens with other languages as well, I was hoping that there would be a way to extract the right characters to begin with and input them into the xhtml.

Thanks!

Toxaris · 05-27-2017, 11:13 AM

Well, you might try OCR, but that will be difficult at well. In most cases it is not an issue, as the keyboard layout in those countries will take care of the correct glyphs.

05-26-2017, 01:20 PM	#1
Nitnesh Junior Member Posts: 2 Karma: 10 Join Date: May 2017 Device: kindle/mobile phones	How to correctly display characters in ePub for non-standard languages? Hi all, I have a PDF as the source file for an ePub in a Ghanaian language and the text that I extract for the ePub does not display the characters/glyphs correctly (see attached screenshots). Embedding the fonts will only help to display the characters in the ePub once I actually have the text, but how can I fix the source text in the xhtml? Also in terms of which fonts to embed (if any?), Times seems to work in other ePubs that I have checked for this language, but I would appreciate any advice that you have with regards to fonts. Google Noto is supposed to work for Ghanaian languages as well, but do any of you have any (similar) experience with font problems and how did you fix them? Thank you in advance! Attached Thumbnails

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
PRS-T3 how to simply display chinese characters in Epub and txt?	sbwtxj	Sony Reader	9	11-14-2013 09:27 AM
Epub cover image does not display correctly	jokapo	Calibre	7	02-26-2012 07:13 AM
Epub does not display images correctly	Amalthia	Calibre	3	09-05-2011 06:44 AM
[Old Thread] Epub cover image does not display correctly	Amalthia	Calibre	30	06-20-2011 11:50 AM
Another ePub does not display correctly.	KjellM	EPUBReader	7	07-16-2010 03:31 AM

05-26-2017, 04:37 PM	#2
Toxaris Wizard Posts: 4,520 Karma: 121692313 Join Date: Oct 2009 Location: Heemskerk, NL Device: PRS-T1, Kobo Touch, Kobo Aura	Since the text files in an ePUB must be UTF-8, the Ghanaia glyphs are probably in there. Just use the correct characters and embed a font that includes these glyphs. You could try Charis Sil, it is rather complete as far as I know. It should not be a problem at all.

05-27-2017, 06:55 AM	#3
Nitnesh Junior Member Posts: 2 Karma: 10 Join Date: May 2017 Device: kindle/mobile phones	Hi Toxaris, Thank you for your reply. You are suggesting to key in the characters in the xhtml, right? This is what I have been doing so far, but it is a very long and strenuous manual job for long books, and prone to errors, as I do not speak the language myself. Also, as I assume that this happens with other languages as well, I was hoping that there would be a way to extract the right characters to begin with and input them into the xhtml. Thanks!

05-27-2017, 11:13 AM	#4
Toxaris Wizard Posts: 4,520 Karma: 121692313 Join Date: Oct 2009 Location: Heemskerk, NL Device: PRS-T1, Kobo Touch, Kobo Aura	Well, you might try OCR, but that will be difficult at well. In most cases it is not an issue, as the keyboard layout in those countries will take care of the correct glyphs.