Dialogflow Basics Theory

What is Dialogflow ?

Dialogflow is Google owned framework that create conversation interfaces for websites, mobile applications, most popular messaging platforms such as Facebook and Twitter and also support voice assistant (Googles Assistant).

The power of Dialogflow lies on Google Cloud Platform services such Google Cloud Speech-to-Text, Cloud Text-to-Speech and Cloud Natural Language. Cloud Speech-to-Text convert audio to text and Cloud Text-to-speech vice versa. Cloud Natural Language uses machine leaning reveal the structure and meaning of text, try demo here

Dialogflow Basic

Agents : a bot that handles conversation with end-users. Dialoflow Agent is similar to a human call center agent.

Intent : A service that a user want from the Agent. End-user send a user-expression by text or voice, Dialogflow will match the user-expression with best intent in the agent. Dialogflow by have two default which are Default Fallback Intent and Default Welcome Intent. Default Fallback Intent is matched when your agent doesn’t user-expression. Default Welcome Intent is triggered whenever the end-user initiate a conversation with your agent.

Responses : An Intent built-in that responses after the intent is matched

Training phrases : An Intent built-in that contain example phrases that end-user might type or say

Entity : Identify and extract useful data from natural language end-user input. The Intent will identify the user needs or request, entity will pick up specific piece of information (currency, date, country etc) that end-user mentioned.

System Entity: Dialogflow pre-built entities that allow agent to extract most popular common concept ( date, street address)

Integrations: Dialogflow offer an opportunity to designer or developer integrate their agent with popular messaging platform like Facebook messenger, twitter and many other third party platform