Skip to content

Latest commit

 

History

History
33 lines (20 loc) · 876 Bytes

README.md

File metadata and controls

33 lines (20 loc) · 876 Bytes
Raw

Dhivehi Reader

Web app to read dhivehi text. Based on tts machine learning models available at https://github.com/DhivehiAI/TTS-Demos. Hope the model gets furthure improvements.

Demo Animation

TODO: Add more supported websites, add support for reading docx, pdf. See How To Add New Scraper

Setup

  • Install required dependencies pip install -r requirements.txt
  • Download pre-trained models see https://github.com/DhivehiAI/TTS-Demos for instructions
  • Extract the model files to ./app/dv_tts/models/

Run the server

Run a test server

python wsgi.py

Run with gunicorn

gunicorn wsgi:app

Deploy to Docker

  • TODO

Deploy to Heroku

Cant deploy to Heroku because the slug size is too big

heroku login
heroku create dhivehi-reader
git push heroku master