AudioConnector Server Reference Guide

Purpose

This repository contains a sample implementation for an AudioConnector Server. This is to be used as a guide to help understand some of the basics of setting up an AudioConnector Server. It is not intended for production purposes. Protocol documentation can be found on the Genesys Developer Portal.

Things to look at to get started

The main session object

The Session class contains methods and logic that handle communicating with the AudioConnector Client.

The ASRService class is responsible for interpreting the incoming audio from the AudioConnector Server. A fake implementation has been provided, and will need to be replaced with an actual ASR engine.

The BotService class is responsible for getting the metadata for a specified Bot, as well as interacting with the Bot itself. For example, this service would be used to match a Bot's response to the input received/interpreted from the ASRService and DTMFService services. A fake implementation has been provided, and will need to be replaced with an actual Bot engine.

The DTMFService class is responsible for interpreting any DTMF digits received from the AudioConnector Client. A base implementation has been provded as a start, but will need to be adjusted to meet any specific requirements for the AudioConnector Server.

The SecretService class is responsible for looking up the secret from a given API Key used during the initial authentication process. A fake implementation has been provided, and will need to be replaced to lookup secrets with whatever service they are stored in.

The TTSService class is responsible for converting text-based responses from the Bot to the appropriate audio to be sent to the AudioConnector Client. A fake implementation has been provided, and will need to be replaces with an actual TTS engine.

Running the server

Requirements

This implementation was written using NodeJS 18.16.0 as a target. If you are using a Node version manager, there is a nvmrc file that specifies this version.

Steps to run the server locally

1. Run npm install in the root of the project.
2. Run npm run start in the root of the project to start the server. The port can be adjusted from within the environment file.