Skip Count is a simple and interactive Alexa skill aimed at making learning to skip count fun. Skill helps reinforce the concept of counting in increments.
Developing a Custom Skill needs understanding of how Alexa Voice Service ( Interaction Schema, Intents, utterance, custom slots ...) works and how to build a service that can expose the custom skill as an API endpoint. Alexa voice service converts the voice command to a JSON and routes the requests to our custom API. Custom API should accept requests from Alexa Voice Service/Skill and send back response complying to the JSON Contract.
Flask-Ask is a plugin for flask, a popular Micro-services framework. Flask-Ask makes building the custom skill API a breeze, it abstracts all the heavy lifting and lets you focus on the core code. Flask-Ask validates requests signature (You can turn this off for Testing), maps requests to view functions, construct response complying with the JSON Interface contract and more.
Configuring Skill is done in Amazon Developer Portal. Here you add Skill Information, Interaction Model (Intent Schema, Custom Slots, Sample utterances), Configuration (Mapping the skill to the custom API). Building the Interaction model is the most critical part of building a custom skill.
- Big Nerd Ranch Tutorial
- Registering Skill in Developer Portal : This step involves configuring skill providing intent schema, sample utterance, custom slots.. and linking to the Flask-Ask API endpoint.
Deploying a flask-ask skill to AWS can be done is so many different ways. IMHO easiest way is to use ZAPPA and deploy the app in a server-less way to AWS Lambda and AWS API Gateway.
- Python Version - 2.7
- Project Structure - Standard python project structure. speech_assets folder has interaction model assets. Check out Structuring Guide
- Nose tests for Unit Testing - tests folder has the test cases. Understanding JSON interface is critical to writing unit tests for flask-ask app.
- Travis CI for build and Code Climate for static code analysis and publishing Test Coverage reports
Clone the repo, cd to the app folder
git clone git@github.com:ravishan16/CountByAlexa.git
cd CountByAlexa
pip install -r requirements.txt
Run app.py launches flask-ask app on http://127.0.0.1:5000/
python countbyalexa/app.py
* Restarting with stat
* Debugger is active!
* Debugger PIN: 145-691-207
* Running on http://127.0.0.1:5000/ (Press CTRL+C to quit)App exposes one endpoint as POST method. Below curl command to invoke launch intent
$ curl -X POST \
http://127.0.0.1:5000/ \
-H 'cache-control: no-cache' \
-H 'content-type: application/json' \
-d '{
"request": {
"locale": "en-US",
"requestId": "amzn1.echo-api.request",
"timestamp": "2017-04-08T21:53:03Z",
"type": "LaunchRequest"
},
"session": {
"application": {
"applicationId": "amzn1.ask.skill"
},
"new": true,
"sessionId": "amzn1.echo-api.session",
"user": {
"userId": "amzn1.ask.account"
}
},
"version": "1.0"
}'- Configuring awscli
- Setup virtualenv,
- Configuring Zappa and runnning deployment
pip install awscli
aws configure
AWS Access Key ID [None]: XXXX
AWS Secret Access Key [None]: XXXX
Default region name [None]: us-east-1
Default output format [None]: jsonvirtualenv venv
venv\Scripts\activate
pip install flask-ask zappa awscli
zappa init
zappa deploy devzappa tail devzappa update devThe following are set as environment variables in Lambda Function:
MY_USERID & MY_NAME are added to personalize the experience only for the alexa userid added to the environment variable.
LOG_LEVEL: LOG_LEVEL is set to defaultINFOthis can be overridden by adding the env variable.MY_USERID: Env variable set to myAlexa User Idthat comes as part of the request. This is optional.MY_NAME: Env variable set to any text\name that will be added as part of the launch intent/ stop intent greeting
-- Ravishankar Sivsasubramaniam