What software/API would you recommend I use to recognize a users voice, transcribe it on the backend, then based on that input initiate a function to play a response from a list of pre-recorded audio files on my server?
I'm looking to build an iOS/Android App that makes the user feel like they are having a very simple conversation with my pre-recorded messages.
I.E user clicks the speak button then says "Hi." Speech recognition recognizes "Hi" then initiates a function to play an audio file response named "Hello".
User says "Tell me a joke" Software recognizes "joke" then plays joke audio response.
If user says something for which there is no valid response then play pre-recorded error message "Sorry didn't get that, try again."
Since I would like this to work on the web, iOS & Android I'm assuming I would mainly need to develop this for the web then from iOS or Android connect to that webpage, do I have that right? How you would go about achieving this?