Please read comments by Ray at
https://github.com/tesseract-ocr/tesseract/issues/654

These talk of the LSTM training process. Since he is retraining, it may be
helpful if you can share info about the new font that you are using so that
he can include it in training.

You can try the fine-tune type training with 4.0 alpha, but seeing the
number of lines required for good training, it maybe best to get the font
included in training by Ray.

Or, you can retrain using jtessboxeditor, with 3.05

- excuse the brevity, sent from mobile

On 24-Jan-2017 2:41 PM, "Sergei Turko" <[email protected]> wrote:

> A little bit better, but I can't use tools for training because of its
> alpha-build version. How can I train tessdata with tesseract 4.0 using apps
> like jTessBoxEditor?
>
> понедельник, 23 января 2017 г., 17:13:14 UTC+3 пользователь shree написал:
>>
>> Have you tried with the 4.0 alpha version of software and new
>> traineddata? What recognition rate do you get with that?
>>
>> ShreeDevi
>> ____________________________________________________________
>> भजन - कीर्तन - आरती @ http://bhajans.ramparivar.com
>>
>> On Mon, Jan 23, 2017 at 1:37 PM, Sergei Turko <[email protected]> wrote:
>>
>>> Hi. I'm trying to train new belarussian font in tesseract 3.04. I edited
>>> box files, made some traineddata files with dictionaries. Now I have
>>> traineddata, which do some correct recognition results, but not 80% quality
>>> at least. How can I improve recognition quality? I have multi tiff with 29
>>> pages. Now I use this traineddata like  "-l bel+bel1"
>>>
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "tesseract-ocr" group.
>>> To unsubscribe from this group and stop receiving emails from it, send
>>> an email to [email protected].
>>> To post to this group, send email to [email protected].
>>> Visit this group at https://groups.google.com/group/tesseract-ocr.
>>> To view this discussion on the web visit https://groups.google.com/d/ms
>>> gid/tesseract-ocr/bf1d3a35-079d-4a79-99b3-9473999e2e9b%40goo
>>> glegroups.com
>>> <https://groups.google.com/d/msgid/tesseract-ocr/bf1d3a35-079d-4a79-99b3-9473999e2e9b%40googlegroups.com?utm_medium=email&utm_source=footer>
>>> .
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>
>> --
> You received this message because you are subscribed to the Google Groups
> "tesseract-ocr" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to [email protected].
> To post to this group, send email to [email protected].
> Visit this group at https://groups.google.com/group/tesseract-ocr.
> To view this discussion on the web visit https://groups.google.com/d/
> msgid/tesseract-ocr/9fedb5f0-d06d-445c-850b-75089e460c2d%
> 40googlegroups.com
> <https://groups.google.com/d/msgid/tesseract-ocr/9fedb5f0-d06d-445c-850b-75089e460c2d%40googlegroups.com?utm_medium=email&utm_source=footer>
> .
> For more options, visit https://groups.google.com/d/optout.
>

-- 
You received this message because you are subscribed to the Google Groups 
"tesseract-ocr" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
To post to this group, send email to [email protected].
Visit this group at https://groups.google.com/group/tesseract-ocr.
To view this discussion on the web visit 
https://groups.google.com/d/msgid/tesseract-ocr/CAG2NduWEf8PsrFjW-vK0RFKJBJKQB2-9wbSDsvO_ycoRcbDSAg%40mail.gmail.com.
For more options, visit https://groups.google.com/d/optout.

Reply via email to