Frontend + Backend
Frontend/ VUI - Voice User Inteface (Alexa Interaction Model) Backend/ Programming Logic (Alexa Lambda Function)
These are the capabilities that aer given by the user to Alexa to perform accordingly or perform some tasks
Alexa enabled device (VUI) -------> Alexa Skill Kit
Alexa, open/play/tell/ask yummy pizza to order large(slot) cheese(slot) pizza
Wake + Launch + Invocation + Utterances (Can have slot)
On asking Alexa to launch specific invocation, it will be further requesting specific INTENT based on request Each INTENT can have multiple utterances and based on these utterances the Alexa gonna train the model and respond
-----------------------
| |
| give me fact |
| a fact | UTTERANCES
| tell me fact |
-----------------------
GetNewFactIntent
---- INTENT ----
We as a developer mostly focus on INTENT and alexa based on intent gonna route the request
Launching a skill
-Slot-
Alexa, ask space facts for a fact about Mars
-wake word- -launch- -invocation name- ------ utterances -----
{planet : 'Mars'}
slotType: slotValue
GetFactByPlanetIntent
------- INTENT -------
-Slot-
Alexa, ask space facts for a fact about {planet}
Mars
Earth
Saturn
Pluto
-wake word- -launch- -invocation name- ------ utterances -----
We can have multiple intents in one Alexa Skill There are 2 types - Custom & Alexa Builtin Intents
Before starting creating Alexa Skill we have to focus on 2 major stuff
VOICE INTERACTION NODEL + PROGRAMMING LOGIC
Amazon developer AWS Lambda/Cloudwatch
(developer.amazon.com) (AWS.amazon.com)
/models(Interaction model) /en-US.json /lambda/custom(Programming logic) /index.js skill.json
TALKING TO ALEXA
|-------- Audio---------------|
| |
| |
User Speech Recognition
| Machine Learning -----Request---> | BACKEND
| NLU <----Response --- |(PROGRAMMING LOGIC)
| Text to Speech
| | AMAZON ALEXA
|-------- Visual---------------|