This is an old question, but IBM Watson is still evolving, so this might be a more relevant answer.
You have 2 options.
You can simply send your application an HTTP REST request (GET or POST), following this tutorial
Or you can use the language SDK.
If you are using nodejs, check out this example .
For java see this example .
Edit
Here is an example git project that I created to integrate text into speech into a simple sample application: conversation-simple-with-text-to-speech
Here's the specific commit that the integration was added to: commit 3564aeb
source share