[Athen] foreign language OCR and TTS
jeffreydell99 at gmail.com
Mon Dec 20 13:19:07 PST 2010
The answer to your questions depend on what foreign languages you are
scanning and reading. For European languages the regular Fine Reader
or ScanSoft OCR engines that come with Kurzweil or Openbook work
great. Because the programs that read the text need the program
displaying the text to identify the language we usually use .doc for
foreign language texts. it is easy to set and change between
languages in MS Word. If you are using a foreign language keyboard
Word will make the change automatically.
For Asian languages we hire a student worker that knows the language
to edit the textbook. for these texts we tried scanning with IRIS
Asian Language edition but the scanned files required so much editing
that it was just easier to have the student worker retype the text.
It is not easy to purchase the voices for many foreign languages
without burchasing it as part of some package with a TTS reading
program. NextUP has a lot of voices availble with their TextAloud
program. Read and Write Gold comes with voices for most European
languages. Freedom Scientific programs have access to a wide variety
of voices with their RealSpeak Direct features. If you want to
purchase quality voices separately NeoSpeech, Acapela, and Cereproc
offer many foreign languages. Most companies that make TTS engines
will only sell development kits to other product developers.
Most US programs will not work with Asian languages very well. JAWS
will read them but will not identify specific details like Hiragana or
Katakana in Japanese. Read and Write Golds features like
Pronunciation tutor and alot of the reading features do not work well
with Asian languages. Some of this info I will be covering with my
CSUN presentation this year on Making Asian Language Materials
Accessible to Students with Print Disabilities.
This is what I can think of off of the top of my head with a head ache
and a desire to start my christmas vacation. If there are more
specific details you need give me a call at 216-875-9938 or e-mail
jeffreydell99 at gmail.com.
On 12/20/10, Maureen Bourbeau <mbourbea at smith.edu> wrote:
> Happy Holidays.........
> I am looking for suggestions and best practices for OCR solutions for
> foreign language reading materials.
> What is preferred OCR software? (for both quality and ease of use)
> what is best format for output file? (.doc, .rtf, .pdf)
> and what text-to-speech tools and voices are being used by the end user?
> In addition to accessibility, our Language Resource Center is beginning to
> explore using foreign language TTS as a support tool to language learning.
> The goal would be to have properly formatted source materials that could be
> used with text reading software.
> Thanks for any and all suggestions!
> Maureen Bourbeau
> Assistive Technology Specialist
> Smith College and Mt Holyoke College
More information about the athen-list