Closed willwade closed 2 years ago
Bit more info: Basically there is a worry that if we use Googles TTS (ResponsiveVoice API) - or any other additional Online API for TTS we will run into two problems:
Also for anything other than spelling all items are currently fixed. I.e. nothing really needs to be dynamically generated. Couldn't we use a third party online TTS tool - but basically run it through a script to create wav recordings of all the words in advance?
Thats the idea at least!
NB: Maybe another issue - but basically using the inbuilt TTS engine on iOS leads to a problem as some voices aren't available. But they are if we use the inbuilt (Nuance system) PLUS allow access to another system (Googles TTS does provide us with a Urdu voice for example which we often need)
Closing in favour of #86
For #161 and other reasons (notably that Bengali voice only available in ResponsiveVoice app - not iOS voices) - that we have some kind of script/tool that builds a mp3 version of a tree in advance (so downloads the voice results from responsiveVoice) - that way we can run on a device with limited TTS support - or for systems where online TTS is a bad idea..