Pipster - Text to audiobook

Convert any text to an audiobook and listen to it from your browser, using Azure Cognitive services.

How it works

Upload any text file and Pipster takes care of the rest. You'll get a link to listen to the audiobook after it's been processed. Since this link is in MP3 format, you can even use it to stream to your own application, like an iOS app.

Note:

If you're developing on Windows, it's highly recommended to use the Docker image, since the paths and dependencies used are in Unix format.

Prerequisites:

1. An Azure Cognitize Services (Speech Services) instance. You can get one here
2. Copy config-example.json to config.json and fill in your own values:
   • subscription-key: Also known as "KEY 1" from your Cognitive Services resource's Keys page
   • resource-region: The region where your Cognitive Services Text to Speech instance lives, i.e. westus2

Setup:

docker build -t pipster .

Run:

docker run -p 5000:5000 pipster

Usage:

1. Upload a file to localhost:5000/convert
2. After waiting a moment, open your browswer to localhost:5000/listen/your-book-name, where "your-book-name" is the name of your file without the extension

End-to-end examples

These use the command line to open a browser to the right place, but you could really just navigate there yourself. These will work once you've completed the prerequisites.

Windows:

docker build -t pipster .
docker run -p 5000:5000 pipster 
curl -F "file=@%cd%\test.txt" localhost:5000/convert
explorer http://localhost:5000/listen/test

OSX:

docker build -t pipster .
docker run -p 5000:5000 pipster &
curl -F "file=@$(pwd)/test.txt" localhost:5000/convert
open http://localhost:5000/listen/test