/voiceplay2

Primary LanguageJavaScriptMIT LicenseMIT

README

Voiceplay is an experimental method of controlling your airplay-connected devices by voice via the Amazon Echo, enabling you to search for and control content from YouTube and your iTunes media library and play it directly on your TV.

Voiceplay is comprised of a node-based server that runs on your local network and two AWS Lambda functions that redirect your voice queries to your home media server. The voiceplayer server communicates with auto-discovered airplay-compatible devices, like an Apple TV. It leverages NaturalNode's naïve Bayes classifier to infer commands based upon natural language.

What can I say

'intents.js' will give you a better idea of what can be said to the Echo. Here are some examples:

  • Ask YouTube to play a music video by Michael Jackson
  • Ask YouTube to play a documentary about Carl Segan
  • Ask iTunes to play Star Trek
  • Ask iTunes give me more infor about this video
  • Launch this video on my mac.
  • Ask YouTube to play something else
  • Ask YouTube / iTunes what are we watching?
  • Ask YouTube / iTunes to pause / repeat / stop.

Requirements

Setup

  • If you haven't already, install node

  • Clone the git repo

  • Run npm install in the project directory

  • Using Finder, Navigate to the folder where voiceplay is installed and open voiceplay.js with your editor of choice to specify its start arguments: your YouTube API key and the path to your iTunes directory.

server.configure({
  iTunesPath: '/Volumes/Media/iTunes/Home Videos/',           // iTunes media library path
  youtubeApiKey: 'AIzaSyDT6ebsYYqQGOY95izP1jtilUsrYdSKXIE',   // example YouTube API key
  port: 8080
});
  • Save the script and with terminal navigate to the directory where you installed the voiceplay module and run it:
node voiceplay
  • If successfully running, you will see:
Amazon Echo / Alexa Voice-Activated Airplay Media Server listening at http://192.168.0.12:8080
Airplay device discovered: Apple TV.  Ready to start receiving requests from the AWS Lambda function.
  • Sign up or log in to the AWS Console and choose Lambda from the Services > All Services menu.
  • Create a Lambda function, skip the intro wizard, name this function 'YouTube' and paste the contents of 'lambda redirect.js'. A description isn't required.
  • Repeat the process for the iTunes media function, naming it iTunes (the names of these functions on the AWS portal isn't important). Take note of each function's ARN endpoint.
  • Seperate from AWS, log in to your Amazon Apps & services Developer Account and under Apps & Services, choose Alexa.
  • You will need to add two Alexa Skills - one for YouTube and the other for iTunes. Each skill has an associated name (youtube | itunes), invocation name (youtube | itunes), intent schema & sample utterances included here that must be specified in order for the functions to work.
  • For each skill, specify the ARN of the lambda functions created on the AWS console.

Notes