Revision 3

This is a major update to STAR which includes a complete user client rewrite and consequently the introduction of several useful features.

• The user client was completely rewritten from scratch in python and WX Widgets, meaning that though feedback must still be gathered to make it look right or even to insure that controls are visible at all, the user client should soon be able to be used without a screen reader within a couple of revisions!
• Due to the script text field being a true richtext control, the previewing hotkeys were changed to control+alt+up, down, and space rather than just control. This was forced on us because NVDA at least seems to always speak when control+up and down are pressed on a text field regardless of any application code.
• It is now possible to press ctrl+alt+enter on any valid speech line in the script to begin auto previewing the entire script up to it's end or the next error.
• You can press alt+backspace anywhere in the main screen to pause and resume any playing speech, thus the stop currently playing speech button was removed from the interface.
• The STAR client can now run all needed components such as the coagulator and providers locally with one click, meaning that STAR should require 0 extra setup assuming you just want to use only the voices on your computer!
• There is now an options dialog with various customizations you can make from the render filename template to the default render output location to the voice preview text and more.
• Now that the options dialog exists, the output device list has been moved to that dialog.
• It is now possible to consolidate the entire script into one audio file! Next to the script field, there is an output subdirectory or consolidated filename field. If you set thhis to a filename such as output.wav instead of a folder name such as output, now a single output.wav file will be created containing all voice clips seperated by a configurable amount of silence.
• The synthesis caching is much improved. Now if you preview a script before rendering it, the render will be almost instant as the cached phrases from the synthesis will now be used while rendering. It is also possible to manually clear the audio cache in the options dialog.
• The render progress sound was removed in favor of a real native progress bar. It sounds a bit less cool but should be visually accessible soon and is far less expensive than playing the sound.
• There are some cool new features in the script syntax, such as selecting only certain lines to render on the minor end to being able to define character aliases per script on the very useful end!
• Typically STAR coagulators now require authentication. To make dealing with this easier, STAR can now save previous hosts you connect to, allowing you to select between them in the options dialog.
• Full documentation is now provided as well as automatic windows client builds on Github.