Baymax

This repository contains a server and client implementation for processing audio files. The server receives audio files from the client, transcribes the audio to text, processes the text using a zero-shot classification model, converts the processed text back to speech, and sends the audio response back to the client which can be useful as a medical assistant. The name of the service is called Baymax, inspired from the movie Big Hero 6

Prerequisites

• Python 3.10 or higher
• pip (Python package installer)

Installation

1. Clone the repository:

git clone https://github.com/Skriller18/Baymax.git

2. Install the required packages:

pip install -r requirements.txt

3. Install the required models and packages for the zero-shot classification model and the speech synthesis model:

Running the Server

1. Navigate to the root directory of the repository.
2. Start the server:

python server.py

Running the client

1. Start the client to record your audio/query that you would like to ask Baymax

python client.py

The client will record audio, send it to the server, receive the processed audio, and play it.

Configuration

1. The server reads instructions from instructions.txt.
2. The client and server IP and port can be configured in client.py and server.py.