Scribe Reader

Building

In order to compile and build this yourself you'll need the tokenizer package. This can be acquired pre compiled from TokenizerMobileWrapper.

Follow the steps in the repository above, downloading the tokenizer.aar and tokenizer-sources.jar file and moving them into the /app/libs/ directory that you'll need to create.

After this you should be able to compile and run the app on your device.

What is this?

This is an app I'm making to help advanced langauge learners consume native Japanese books and content. This takes heavy inspiration from Duolingo with it's gamified quizes throughout the reading.

Core Features

• In app parsing of .epub file removing unneccessary text
• In app tokenization of Japanese text
• On device dictionary using JMDict (custom dictionary support planned no eta)
• Searchable dictionary from anywhere in the app with sentence tokenization and custom input
• Basic TTS to read out the story
• SRS/Flash card system
• Add cards to review when definition/reading is viewed during story
• Song Lyric Retrieval and analysis
• Anime/Video Subtitle processing/retrieval
• Website retrieval, processing and tokenization

Start of Chapter 1 of a certain light novel

Example of UI element for viewing the definition of a word with ambiguity

Example of UI Element for viewing the definition of a word without ambiguity

Example of UI Element for viewing the reading, dictionary form, and part of speech of a word

As of initial commit to this repository, epubs are no longer processed in another application and retrieved by the app. Now all is done in the app.

Attributions

This application uses JMDict

This repository contains a few other slightly altered repositories because I couldn't figure out how to add them through gradle

• TokenizerMobileWrapper / Kogame - for tokenization of Japanese text
• Epublib - for parsing .epub files
• Jazzlib - used by Epublib for unzipping files