koa-pig is JavaScript application, that let you to synthesize voice from text. Based on Amazon Polly. Developed to run on ARM processors systems, also can be run on linux.
clone repo:
$ git clone git@github.com:Shkurpylo/koa-pig.git
$ cd koa-pig
install packages with npm :
$ npm install -g concurrently
$ npm install -g pm2
$ npm install
or with yarn:
$ yarn global add concurrently
$ yarn global add pm2
$ yarn install
for better performance recomendated to install mpg123 player: $ sudo apt-get install mpg123
if you wan't to install mpg123 - set another player in player.json "player" field, (for example "aplay" - default player on RaspberryPi) or set an empty string - in that case will be used omxplayer (default on ubuntu).
!important set your Amazon accessKeyId and secretAccessKey in aws_config.json
for run application:
$ npm run dev
for run application in background:
$ npm run deamon
api runs on port 3005
routes: GET /voices - get voices list POST /say - synthesize and play speach from text.
- Amazon Polly - Turn text into lifelike speech using deep learning
- Koa - Expressive middleware for node.js