About | Demo | Installation | Running the App | Authors

Abstract

Talk-a-Palooza is an innovative app designed to address the need for effective language development tools for children.
It features a machine learning-powered recommendation system that offers users personalized content for vocabulary expansion and assesses pronunciation accuracy during engaging and fun audio exercises.
With a user-friendly interface for both parents and children, Talkapalooza empowers young learners on their language development journey.

Built with:

 

Demo

 

👀 Watch Demo

InitialAssessment.mp4

AudioExercise_v1.0.mp4

Dashboard.mp4

   

Installation

1. Setting Up The Correct .env File:

To ensure that our credentials are not exposed on Github, we have set up a .env file containing necessary data to connect to our MongoDB Atlas database.

The .env file contains all sensitive data necessary to run the app, including database credentials in the following format:

#MONGO CONFIG
MONGO_URI=mongodb+srv://<user>:<password>@<cluster>.aj9zhtw.mongodb.net/<user>?retryWrites=true&w=majority

Therefore, to be able to run the app, please contact Rose Jeantet to receive the necessary credentials.

 

2. Python Setup:

To receive remedy recommendation from the Data Science API, Python 3 needs to be installed. We used Python3.8 with FastAPI, so we recommend choosing the same version.

• Download Python for Windows and Installation Docs for Windows
• Python for Mac and Installation Docs for Mac

 

3. Clone Into The Git Repository:

To pull this application to your local machine, execute the following commands from your command line:

git clone https://github.com/TechLabs-Berlin/ss23-talk-a-palooza.git

 

Install dependencies:

To run the app, you first need to install all required dependencies for the client, and servers (node and python)

 

Install the client server (React)

Move to the newly cloned project directory, navigate to the client folder, and install the dependencies

cd ss23-talk-a-palooza
cd app/client

npm i

Install the backend server (Node)

Navigate to the server folder, then install the dependencies:

cd ..
cd server

npm i

Install the deep learning server (Python)

Navigate to the dl_api folder, and install Python dependencies

cd ..
cd fastapi/dl_api

python3.9 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

Install the data science server (Python)

There is two DS api available, you need to navigate to the ds_api folder, and install Python dependencies in both folders "soniaapi" and "ticianeapi"

cd ..
cd ds_api/soniaapi

python3.9 -m venv venv
source venv/bin/activate
pip install -r requirements.txt

cd ..
cd ticianeapi

python3 -m venv .venv
source .venv/bin/activate
python3 -m pip install .

// Option to install via Docker (cf. README in Ticiane's folder)

   

Running the App

Navigate to the client folder to start the app (serving port 19006)

cd ..
cd client

npm run web

Navigate to the server folder to start the local server (serving port 3001)

cd ..
cd server

npm run dev

To enable the speech recognition tool, navigate to the dl_api folder to start the local server - FastAPI (serving port 8000)

cd ..
cd fastapi/dl_api

uvicorn main:app --reload

To enable the machine learning recommendation tool, navigate to the ds_api folder to start the local server - FastAPI (serving port 8001)

cd ..
cd fastapi/ds_api/soniapi

uvicorn mainy:app --host 0.0.0.0 --port 8001

cd ..
cd ticianeapi

python3 -m ss23_talk_a_palooza

// Option to run via Docker (cf. README in Ticiane's folder)

   

Authors

User Experience:   Camila Canteros Lopez  

WD (fullstack):       Rose Jeantet  

Data Science:         Sonia Lyn, Ticiane Rover  

Deep Learning:      Aljoscha Beiers  

 

Mentors

Bogdan Ciobotaru,  Christoph Dansard