This page will help you get started with Speech API.
Speech is an intermediary step in transforming your Script into the complete, high quality audio file and it's responsible for the text-to-speech part.
Any voice available in the Voice Library can be used to synthesize speech per Script section.
It is necessary to produce Speech for your Script before moving on to Mastering.
Handy notes
Synthesizing long scripts may exceed the 30 second response limit!
In such case, a synchronous POST /speech call will return a requestId after 30 seconds, while the processing continues in the background. Make a GET /speech?requestId=(...) call to check the state of processing.
Think of Script sections as of building blocks
Each section is processed separately, with the possibility of using a different voice, speed, and parameters.
Hence, the more sections has your script, the more processing capabilities you have (and it's faster!).