dinubs / chispas-ai

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

✨ Chispas

Overview

The Language Learning App aims to provide an interactive and adaptive learning experience for individuals looking to improve their language skills. Utilizing OpenAI's ChatGPT for linguistic analysis and tailored educational content, the application offers a multi-faceted approach to language learning.

image

Core Features:

  • Random Text Blocks: The app will provide the user with a block of text in the language they are trying to learn.
  • Word Selection: Users can select or type in words that they find difficult or would like to improve upon.
  • Database Storage: These selected "difficult" words are stored in a user-specific database.
  • Thematic Analysis: ChatGPT will analyze these words to determine common themes (e.g., past tense verbs, food-related vocabulary).
  • Detailed Explanations: The app will provide detailed explanations and examples for each unknown word and its related themes.
  • New Examples: Users will then be provided with new example sentences or paragraphs incorporating these "difficult" words.
  • Progression: Finally, new text blocks, tailored to each user's progress, will be introduced.

This process will then, theoretically, be applied recursively to progressively teach people a langue in chunks that are related to each other.

CI

Setup

Setup env

scripts/setup

Generate an OpenAI token https://platform.openai.com/account/api-keys and supply it as an environment variable, OPENAI_API_KEY. Or copy/edit the example .env file.

cp .env.example .env
vim .env
rm -f language_learning_app.db
pipenv run cli init-db
pipenv run cli serve
# open new tab
curl -X POST -d "username=chispas&password=chispas" http://127.0.0.1:5000/api/users/add

Run `pipenv run cli secret-key` to generate a new secret key.

Use the cli to run the app

```bash
pipenv run cli serve

Can reset words table with pipenv run cli init-db.

Then visit http://localhost:5000/.

Test

pipenv run test

About


Languages

Language:Python 79.3%Language:HTML 13.3%Language:CSS 4.2%Language:Shell 3.2%Language:JavaScript 0.1%