Autogenerate Graded Readers

FOR Text-To-Speech you will need an Azure account setup and put your API key and region in constants.py

For an Azure account go HERE
Create a Speech service HERE
From there you should be able to grab the API key and region to put in constants.py

Usage：

virtualenv venv
pip3 install -r requirements.txt
python3 story_to_tex.py input_file.txt

This will generate a few things

1. LaTeX file for creating the PDF document
2. A text file with just the chinese (for personal TTS use)
3. A text file with chinese-english-chinese (for personal TTS use)
4. Audio file of just chinese audio
5. Audio file of the chinese-english-chinese

To compile the .text file into a formatted PDF you will need some form of LaTeX installed. I recommend using miktex, which you can download HERE

Text input format

Test input files are in the following form

中文
chinese

(刘备)
Liu Bei

[朋友][friend]
friend

\p一二三
one two three

where we have lines in order of chinese, english, blank. Then there are some additional formatting:

• parenthesis for bolding (names)
• square brackets for vocab [中文][chinese]
• \p for starting new paragraphs