ChatBot made with Angular and Python It uses the library kokoro (https://huggingface.co/hexgrad/Kokoro-82M) and whisper-large-v3-turbo (https://huggingface.co/openai/whisper-large-v3-turbo)

It has been tested in Ubuntu 24.04 with Python 3.9.21.

Executing with docker

docker run -it -p5000:5000 --gpus all -e GOOGLE_API_KEY=$GOOGLE_API_KEY cuda-python39-image IMPORTANT: You will need a good Graphic Card ,16GB memory and 10 GB or more of hard disk to install this app.

Back

Dependencies

The back is in the directory 'back':

With conda installed, create the needed: environment

conda env create -f environment.yml

This will take a while because it has a lot of dependencies.

Update your environment if you have one installed

conda env update --file environment.yml

Important: You will need to install the next packages in your linux.

sudo apt install ffmpeg sudo apt-get -qq -y install espeak-ng > /dev/null 2>&1

Create robotito_db directory to save data. In the directory back (outside src) execute:

mkdir robotito_db

You must have your key to interact with the api of gemini. You can create one in https://ai.google.dev/gemini-api/docs/api-key

Create this environment variable:

GOOGLE_API_KEY="<YOUR_KEY">

Start back

Execute:

python src/flask_server.py

Execute docker of postgres manually

Example:

docker run --name robotito_db -p5432:5432 -e POSTGRES_USER=robotito -e POSTGRES_PASSWORD=secret -d postgres

Front:

You have the front in the directory 'front':

It's made with Angular so you will need node 19:

Dependencies

Install dependencies with:

npm install

Start Front:

npm start Start using 'start.sh'

Example of context:

You are going to pretend that you are my English teacher. You can only speak English. It is very important that you correct me if I make any mistakes or if my answer does not make sense or it doesn't sound natural. Don't be overly helpful, the idea is having a conversation so you have to make the conversation flow. Remember, the most importance in our conversation is tell me if my sentence was correct.

Make DockerFiles

FRONT

cd front ng build docker build -t robotito_front . docker tag robotito_front chuchip/robotito_front:latest docker push chuchip/robotito_front:latest

BACK LIGHT

cd back docker build -t robotito_back -f Dockerfile-light . docker tag robotito_back chuchip/robotito_back:latest docker push chuchip/robotito_back:latest

BACK FULL

cd  # go to home directory
docker build -t robotito_back .