Learning with Texts

Learning with Texts (LWT) is a tool for language learning by reading. It is a self-hosted web application.

You feel that you won't learn much by translating dumb sentences or using grammar books? Learning With Texts offers you the possibility to learn by reading texts in your target language. Concept: when you don't know a word in a text, just click it. We show you the translation, and you will have regular tests to remember it. Ready to go?

Important

THIS IS A THIRD PARTY VERSION. This version is not the official one, and brings many improvements and new features. It is quicker, has smaller database size, and is open for contributions. The official version is on source forge

Note

HugoFara: I don't plan to continue developping LWT on a regular basis. While I may keep maintaining for fun, I recommend jzohrab/lute-v3 as the main target for development effort in the LWT software family.

Installation

As LWT is self-hosted, you will need a server, which can be your computer. You can either use Docker (recommended), or install it on your machine.

Docker (any OS)

Install Docker (if not already done).

• For an light-weight installation, you may use HugoFara/lwt-docker-installer.

• To build from source, download the latest release and run:

  cd lwt
  docker compose up  # Now open http://localhost:8010/lwt/ in a browser

Linux

1. Get the latest GitHub release. You can also try to download the latest stable version if you want the cutting-edge updates (that may include some bugs).
2. Start a shell in the downloaded folder an run: ./INSTALL.sh. You may need to run chmod +x ./INSTALL.sh first.

Other Systems

1. Please follow: docs/install.md for setup instructions.
2. Create connect.inc.php with an existing database user. Everything is explained at docs/info.html.

And you are ready to go!

Description

LWT is a language learning web application. To learn a language, you need to practice, and we guide you in reading exercises.

First copy/paste any text you want to read. It can be raw text or an RSS feed.

Then, we parse the text. Unknown words will be displayed with different colors, just click them to see it in a dictionary.

Read as much as you want!

To make sure you memorize new words, you can take review exercises.

The difference with popular remembering software like Anki is that we keep track of the context to help you. By the way, we also ship an Anki exporter.

Features

Full features list: docs/features.md

Features included from the official LWT software:

• Support for almost 40 languages.
• Text parsing for roman languages, right-to-left, and East-Asian ideogram systems
• Translate words on-the-fly
• Add an audio track and read it online
• Practice words you don't remember
• Statistics to record your progress

Features not in the official LWT

Full new features list: docs/newfeatures.md

Features that were added by the community:

• Support for mobile
• Automatically import texts from RSS feeds
• Support for different themes
• Display translations of terms with status in the reading frame
• Multiwords selection (click and hold on a word → move to another word → release mouse button)
• Bulk translate new words in the reading frame
• Text to speech
• Optional "ignore all" button in read texts
• Key bindings in the reading frame
• Selecting terms according to a text tag
• Two database backup modes (new or old structure)

Improvements compared to the official LWT

• Database improvements (db size is much smaller now)
• Longer (>9) expressions can now be saved (up to 250 characters)
• Save text/audio position in the reading frame
• Google api (use 'ggl.php' instead of '*http://translate.google.com' for Google Translate)
• Improved Search/Query for Words/Texts
• Term import with more options (i.e.: combine translations, multiple tag import)
• Support for MeCab for Japanese word-by-word automatic translation.
• You can include video files from popular video platforms.
• Code documentation.
• Code is well organised, making debugging and contribution easier.

Contribute

Complete explanation: docs/contribute.md

To contribute, you need to clone or fork this repository, and Composer. The composer package is at hugofara/lwt.

Run git clone https://github.com/HugoFara/lwt

Next, got to the lwt folder and use composer install --dev.

In short:

git clone https://github.com/HugoFara/lwt
cd lwt
composer install --dev

Branches

• The stable branch is master. Last commit on this branch is considered to be bug-free.
• The dev branch is for unstable versions.
• The official branch is for the official LWT Releases. Any other branch if considered under development.

Useful links

• General documentation at docs/info.html.
• Please find more help at docs/index.html.
• You can also contact the community using GitHub or Discord.

Alternatives

See also: docs/links.md

• jzohrab/LUTE is a rewrite of LWT with modern tools such as Symfony.
• FLTR ◆ Foreign Language Text Reader, a Java clone, by lang-learn-guy (original author of LWT), it is a standalone installation.
• simjanos-dev/LinguaCafe: a beautiful equivalent in Vue.js and PHP.

Unlicense

Under unlicense, view UNLICENSE.md, please look at http://unlicense.org/.

Let's learn new languages!