• DOI:
• Build status:
• Code coverage:

Source URL: https://github.com/borh/jsrs

Hosted instances:

• https://cuckoo.js.ila.titech.ac.jp/app/
• https://cuckoo.js.ila.titech.ac.jp/c-jsrs/

The Japanese Speech Rating System (JSRS) is a Python (Django) server and client application that crowdsources preference ratings of Japanese learner speech. Raters listen to two consequitive readings of the same sentence spoken by non-native and native Japanese speakers and indicate their preference between them.

The system also includes a sophisticated analysis module powered by the mdpref R package developed by Okubo & Mayekawa (2015) available here.

Currently, audio files and associated metadata must be imported from a spreadsheet. Dataset import files (Django fixtures) are generated using the provided generate_fixtures.py script contained in the project's root directory.

Audio files must be place under audio and have the following directory and naming structure:

WAV files will be automatically converted to MP3 format. All files will also be normalized with mp3gain.

You will need to create an Excel file named データID.xlsx containing the following structure:

First sheet containing the set of sentences with number corresponding to the number used in the audio file name, order the priority of the sentence, and sentence text showing the original text stimulus:

Second sheet containing:

Third sheet containing:

General settings are inherited from the Cookiecutter Django template. See the settings documentation there.

source venv/bin/activate # or activate.fish/etc. depending on your shell

pip install -r requirements/local.txt
pip install -r requirements/production.txt
pip install -r requirements/test.txt

Install needed R pacakges (see README):

cd R
curl -O http://mayekawa.in.coocan.jp/Rpackages/lazy.mat_0.1.3.zip
curl -O http://mayekawa.in.coocan.jp/Rpackages/lazy.tools_0.1.3.zip
curl -O http://mayekawa.in.coocan.jp/Rpackages/lazy.mdpref_0.1.2.zip
unzip -x lazy.mat_0.1.3.zip
unzip -x lazy.tools_0.1.3.zip
unzip -x lazy.mdpref_0.1.2.zip
Rscript -e "install.packages('naturalsort')"
R CMD INSTALL lazy.mat
R CMD INSTALL lazy.tools
R CMD INSTALL lazy.mdpref
Rscript -e "source(\"http://bioconductor.org/biocLite.R\"); biocLite(\"pcaMethods\")"
cd ..

Create the database and run schema migrations:

createdb -U postgres jsrs # -U <user-running-jsrs-app> would be better
# 'sudo -u postgres createdb -U postgres jsrs' under peer auth
# Replace as necessary:
export DATABASE_URL="postgres://<pg_user_name>:<pg_user_password>@127.0.0.1:5432/jsrs"
python manage.py migrate

Alternatively, if importing an existing database dump jsrs-2017-07-19.sql into jsrs-dbname:

createdb -T template0 -U postgres jsrs-dbname
psql -U postgres jsrs-dbname < jsrs-2017-07-19.sql

Look at env.example and adjust accordingly. Then, load the environment variables into the environment before running the app.

./gunicorn.sh

NOTE: Currenly only production settings are supported.

python manage.py runserver

Visit localhost:8000 in your browser to check if everything worked.

Optionally setting up MailHog as SMTP app:

curl -O https://github.com/mailhog/MailHog/releases/download/v0.2.1/MailHog_linux_amd64
mv MailHog_linux_amd64 mailhog
chmod +x mailhog
export DJANGO_EMAIL_BACKEND=./mailhog

Install npm prerequisites and confirm serving with grunt works:

npm install
grunt serve

Visit localhost:8000 to check if everything worked.

First generate audio fixtures using provided python script. Then load the fixtures into the database.

Audio files should be accessible from media/.

./generate_fixtures.py
python manage.py loaddata audio-sentence-fixtures.json
python manage.py loaddata audio-reader-fixtures.json
python manage.py loaddata audio-audio-fixtures.json

To create a normal user account, just go to Sign Up and fill out the form. Once you submit it, you'll see a "Verify Your E-mail Address" page. Go to your console to see a simulated email verification message. Copy the link into your browser. Now the user's email should be verified and ready to go.

To create an superuser account, use this command:

$ python manage.py createsuperuser

For convenience, you can keep your normal user logged in on Chrome and your superuser logged in on Firefox (or similar), so that you can see how the site behaves for both kinds of users.

python manage.py makemigrations
python manage.py migrate

To run the tests, check your test coverage, and generate an HTML coverage report:

$ coverage run manage.py test
$ coverage html
$ open htmlcov/index.html

Moved to Live reloading and SASS compilation.

It's time to write the code!!!

N.B. The integration tests will not run on Windows.

To install the test runner:

$ pip install hitch

To run the tests, enter the jsrs/tests directory and run the following commands:

$ hitch init

Then run the stub test:

$ hitch test stub.test

This will download and compile python, postgres and redis and install all python requirements so the first time it runs it may take a while.

Subsequent test runs will be much quicker.

The testing framework runs Django, Celery (if enabled), Postgres, HitchSMTP (a mock SMTP server), Firefox/Selenium and Redis.

We providing tools and instructions for deploying using Docker and Heroku.

See detailed cookiecutter-django Heroku documentation.

See detailed cookiecutter-django Docker documentation.

The steps below are only for the record, do not run them after initial project setup.

First set up a Python 3 virtual environment:

pyvenv venv # alternatively: virtualenv-3.6 venv
source venv/bin/activate # or activate.fish/etc. depending on your shell
pip install --upgrade pip

Use cookiecutter to generate project:

pip install --upgrade cookiecutter
cookiecutter https://github.com/pydanny/cookiecutter-django.git

This project is licensed under the MIT license.