Text to Speech Using Google's Wavenet Model

This script can be run locally on Mac or Linux OS to transpose a text file "text.txt" into a series of sequential dictated audio files. A great use case it to transpose pdf's documents into audio books.

Create a Service Account

Download Node.js

Linux

sudo apt-get install node sudo apt-get install npm

Download dependencies

From the command line navigate into the Wavenet-Text-to-Speech-Example and run the following command. npm install

Execute Script

Swap out the location of your service account json key file. GOOGLE_APPLICATION_CREDENTIALS="/Users/refactored/Desktop/wavenet-service-account.json" npm start

The above command allows us to bypass having to download the gcloud SDK utility by directly passing our service account credentials.

Alternative way to run the script GOOGLE_APPLICATION_CREDENTIALS="/Users/refactored/Desktop/wavenet-service-account.json" node main.js

YouTube Step by Step Tutorial

https://www.youtube.com/watch?v=uxjPTalCCT0