This is a proof-of-concept program to do text-to-speech using the Go AST parser. It attempts to do an English-language rendering of what the code is doing, rather than just reading individual words and characters as a regular TTS engine would.
It currently relies on the Mac OSX "say" program, so it won't run on another platform without reworking the speak function.
To run:
go get github.com/gospeak/...
saygo [options] gofiles
- -q option to disable the speaking if you are just debugging the language processing.
- -func funcname to only read out a specific function
- -o anAudioFile.aiff to save the speech to a file
Otherwise, just specify Go files on the command-line and it will read out each one.
I restructured the gospeak API so that it passes data around with the calls to facilitate multiple instances. If for some reason you still need the previous API, I tagged it as 0.0.1. The saygo command still works the same way.
I also added a SpeakGoString call that makes it easier to write unit tests.