At AWS Summit 18-19 April 2017 there was a workshop for building Alexa Skills, which was the first hands-on Alexa workshop I'd been to, so I decided to give it a go.

What it does

Top 100 singles is the first-ever Alexa Skill to tell us what the Top 100 best selling singles are. It includes information on the singer, the record company, the year the song was popular and how many records were sold.

How I built it

This app was built with Amazon Alexa, Amazon Lambda, Amazon Cloudwatch to track and troubleshoot log files, Node.JS (the Alexa sdk) and tested with

Challenges I ran into

Originally, I wanted Alexa just to read through the Wikipedia tables of best selling singles but that didn't quite work so I had to do the app differently.

There was also data I wanted to include, such as the record label, which Wikipedia didn't have on their data tables so I had to collect that data from scratch.

I also learnt that Alexa doesn't accept Intents that are numbers such as 100 so I changed it to "One hundred".

Accomplishments that I'm proud of

12,500+ Alexa Skills have been certified in the last two years and this is the first-ever "Top 100 singles"!

Most Alexa apps seem to have been coded by male engineers and I'm happy I was there in the AWS Summit workshop to represent female engineers and to learn the technical skills I needed. Male:female ratio in the workshop seemed to be 120:3.

What I learned

The format of Wikipedia and its data tables would need to be changed to be more friendly for voice apps.

What's next for Twain's Alexa Skills Development

I'm planning to do a Chocolate Recommendations app whereby Alexa then enables the buyer to order chocolate from Amazon by voice. So if I win a prize in this hackathon that would tool me more to do that since I don't own any Alexa devices!

More broadly, I'm hoping to integrate Alexa, Polly, Lex and other technologies so we can leave "voice crumbs" of our emotional experiences for friends to discover on a visual map.

"Voice crumbs" [(C) Twain 2017. All rights reserved. Coined by Twain, weekend of 21- 23 April 2017.]

Built With

Share this project: