---
name: "combine_lang_model"
section: "1"
platform: "linux"
package: "tesseract-ocr"
version: "5.5.0-1+b1"
release: "Debian 13 (Trixie)"
platforms: ["linux"]
url: "https://new.carta.tech/combine_lang_model"
---

# combine\_lang\_model(1)

generate starter traineddata

## NAME

combine\_lang\_model - generate starter traineddata

## SYNOPSIS

**combine\_lang\_model** --input\_unicharset *filename* --script\_dir *dirname* --output\_dir *rootdir* --lang *lang* \[--lang\_is\_rtl\] \[pass\_through\_recoder\] \[--words file --puncs file --numbers file\]

## DESCRIPTION

[combine\_lang\_model(1)](/combine_lang_model/1.md) generates a starter traineddata file that can be used to train an LSTM-based neural network model. It takes as input a unicharset and an optional set of wordlists. It eliminates the need to run [set\_unicharset\_properties(1)](/set_unicharset_properties/1.md), [wordlist2dawg(1)](/wordlist2dawg/1.md), some non-existent binary to generate the recoder (unicode compressor), and finally [combine\_tessdata(1)](/combine_tessdata/1.md).

## OPTIONS

- *--lang lang*

  The language to use. Tesseract uses 3-character ISO 639-2 language codes. (See LANGUAGES)

- *--script\_dir PATH*

  Directory name for input script unicharsets. It should point to the location of langdata (github repo) directory. (type:string default:)

- *--input\_unicharset FILE*

  Unicharset to complete and use in encoding. It can be a hand-created file with incomplete fields. Its basic and script properties will be set before it is used. (type:string default:)

- *--lang\_is\_rtl BOOL*

  True if language being processed is written right-to-left (eg Arabic/Hebrew). (type:bool default:false)

- *--pass\_through\_recoder BOOL*

  If true, the recoder is a simple pass-through of the unicharset. Otherwise, potentially a compression of it by encoding Hangul in Jamos, decomposing multi-unicode symbols into sequences of unicodes, and encoding Han using the data in the radical\_table\_data, which must be the content of the file: langdata/radical-stroke.txt. (type:bool default:false)

- *--version\_str STRING*

  An arbitrary version label to add to traineddata file (type:string default:)

- *--words FILE*

  (Optional) File listing words to use for the system dictionary (type:string default:)

- *--numbers FILE*

  (Optional) File listing number patterns (type:string default:)

- *--puncs FILE*

  (Optional) File listing punctuation patterns. The words/puncs/numbers lists may be all empty. If any are non-empty then puncs must be non-empty. (type:string default:)

- *--output\_dir PATH*

  Root directory for output files. Output files will be written to \<output\_dir\>/\<lang\>/\<lang\>.\* (type:string default:)

## HISTORY

[combine\_lang\_model(1)](/combine_lang_model/1.md) was first made available for tesseract4.00.00alpha.

## RESOURCES

Main web site: **https://github.com/tesseract-ocr** Information on training tesseract LSTM: **https://tesseract-ocr.github.io/tessdoc/TrainingTesseract-4.00.html**

## SEE ALSO

[tesseract(1)](/tesseract/1.md)

## COPYING

Copyright (C) 2012 Google, Inc. Licensed under the Apache License, Version 2.0

## AUTHOR

The Tesseract OCR engine was written by Ray Smith and his research groups at Hewlett Packard (1985-1995) and Google (2006-2018).