Make a "higher-level" SSIP library. #73
Labels
enhancement
New feature or request
help wanted
Extra attention is needed
TTS
Improvements to the text to speech subsystem
The currently
ssip-cleint
library is sufficient to handle all communication withspeech-dispatcher
, however, I'm somewhat concerned that there are a lot of "weird" things about it that won't make sense unless you're familiar with the protocol. For example: If you want to send speech, you must await the return value, or when you send next, speech-dispatcher will have issues, you need to use SendLines(Vec) instead of SendLine(String), unless you want to add your own final "." line to end the wrong.There should be a less obtuse way to communicate with speech-dispatcher. I'm proposing something like
ssip
orspeechd
(on crates.io).This library will have functions (async and non-async) to communicate with speech-dispatcher, and commands like "cancel", "stop", or "quit" will jump the lin in the internal queue.
Not sure exactly how we'll implement all of this, but let's put some ideas in here.
The text was updated successfully, but these errors were encountered: