self-hosted-cookbook/apps/other/language-tool.md

LanguageTool

Multilingual grammar, style, and spell checker (similar to Grammarly).

• Official Site
• Github repo
• Docker Hub

Optional n-gram data

LanguageTool can make use of large n-gram data sets to detect errors with words that are often confused, like their and there.

1. Download & unzip n-gram data for your language: https://languagetool.org/download/ngram-data/

   wget https://languagetool.org/download/ngram-data/ngrams-en-20150817.zip
   unzip ngrams-en-20150817.zip
   

2. Uncomment n-gram volume and env variable below

docker-compose.yml

---
version: "3"
services:
  languagetool:
    image: erikvl87/languagetool
    container_name: languagetool
    restart: unless-stopped
    ports:
        - 8010:8010  # Using default port from the image
    environment:
        # - langtool_languageModel=/ngrams  # OPTIONAL: Using ngrams data
        - Java_Xms=512m  # OPTIONAL: Setting a minimal Java heap size of 512 mib
        - Java_Xmx=1g  # OPTIONAL: Setting a maximum Java heap size of 1 Gib
    volumes:
        - ./ngrams:/ngrams

And then, in a browser extension:

• in LanguageTool API server URL: select Other server
• enter: http://<SERVER IP>:8010/v2

Alternative docker-compose.yml

This is using libregrammar version, which (allegedly) provides the premium features for free.

• Github repo for libregrammar
• Github repo for the docker image

---
services:
  languagetool:
    image: registry.gitlab.com/py_crash/docker-libregrammar
    container_name: languagetool
    restart: unless-stopped
    environment:
      - langtool_languageModel=/ngrams  # OPTIONAL: Using ngrams data
      - Java_Xms=512m  # OPTIONAL: Setting a minimal Java heap size of 512 mib
      - Java_Xmx=1g  # OPTIONAL: Setting a maximum Java heap size of 1 Gib
    ports:
      - "8081:8081"
    volumes:
      - ./ngrams:/ngrams