NOVA API

This is an Express API that provides various endpoints to build applications that can generate text completion and speech synthesis using the OpenAI's GPT-3 model and the VoiceRSS API.

Usage

The API is hosted on Heroku. So you can use it without having to install it locally.

Base URL: https://api-nova-voicebot.herokuapp.com

Here's a list of all the available endpoints:

Welcome

• GET /

This endpoint returns a welcome message to confirm that the API is up and running.

Request

GET /

Response

Welcome to the NOVA API!

Completion

• POST /api/completion

  This endpoint takes in a request body containing a text prompt and returns the completion of that prompt using the OpenAI's GPT-3 model.

Request

POST /api/completion
Content-Type: application/json
Authorization: Bearer <token>

{
    "prompt": string;
    "user": string;
}

Response

Accept: application/json

{
    "success": true;
    "completion": {
        "id": string;
        "object": "text_completion",
        "created": number;
        "model": "text-davinci-003",
        "choices": [
            {
              "text": string; # The actual completion
              "index": 0,
              "logprobs": null,
              "finish_reason": string;
            }
        ],
        "usage": {
            "prompt_tokens": number;
            "completion_tokens": number;
            "total_tokens": number;
        }
    }
}

Speech

• POST /api/speech

  This endpoint takes in a request body containing a text input and returns the speech synthesis of that text.

Request

POST /api/speech
Content-Type: application/json
Authorization: Bearer <token>

{
    "text": string;
    "lang"?: string;   # Default: "en-US"
    "speed"?: string;  # Default: "0"
    "codec"?: string;  # Default: "mp3"
    "format"?: string; # Default: "8khz_8bit_mono"
    "b64"?: string;    # Default: "true"
}

Response

Accept: application/json

{
    "success": true;
    "speech": string; # "Base64-encoded audio"
}

User

• POST /user/signup

  This endpoint creates a new user account.

Request

POST /user/signup
Content-Type: application/json

{
    "email": string;
    "password": string;
}

Response

Accept: application/json

{
    "success": true;
    "user": {
        "username": string;
        "email": string;
    }
}

• POST /user/login

  This endpoint logs in an existing user.

Request

POST /user/login
Content-Type: application/json

{
    "email": string;
    "password": string;
}

Response

Accept: application/json

{
    "success": true;
    "token": string; # User token, expires in 1 day
}

• POST /user/userinfo

  This endpoint returns user information based on the token provided in the request.

Request

POST /user/userinfo
Content-Type: application/json

{
    "token": string; # User token
}

Response

Accept: application/json

{
    "success": true;
    "user": {
        "username": string;
        "email": string;
    }
}

• POST /user/delete

  This endpoint deletes a user account based on the token provided in the request.

Request

POST /user/delete
Content-Type: application/json

{
    "token": string;
}

Response

Accept: application/json

{
    "success": true;
}

Support

• POST /support/contact

  This endpoint sends an email to the NOVA support team.

Request

POST /support/contact
Content-Type: application/json

{
    "name": string;
    "email": string;
    "subject": string;
    "message": string;
}

Response

Accept: application/json

{
    "success": true;
    "message": "Message sent successfully";
    "info": SMTPTransport.SentMessageInfo; # See https://nodemailer.com/message/
}

Errors

All errors are returned in the following format:

Status: <status code>

{
    "success": false;
    "error": string;
    "fields"?: string[];
    "categories"?: string[]; # Only for /api/completion, when the prompt is against OpenAI's guidelines
}

Installation

Prerequisites

• Node.js
• npm or yarn
• MongoDB
• OpenAI API key
• VoiceRSS API key
• Gmail account

Setup

1. Clone the repository

git clone https://github.com/YacineSteeve/nova-voicebot-api.git

2. Install dependencies

yarn install

3. Add environment variables:

• Create a .env file in the root directory
• add the environment variables listed in the .env.example file.

Note: NODE_ENV should be set to development.

4. Start the server

yarn dev

License

MIT

Author

Yacine BOUKARI