I am building a alexa app for hotel/food ordering system. Here i have defined a custom intent with custom slot and many utterances for the same.
There are infinite ways a user can utter for his order, and we can't define all the utterances in the schema to serve the user. So i am trying build my own NLU engine, to handle all other cases in which i don't get the desired slots value.
I want amazon alexa to provide full user utterance for every conversation whether alexa recognizes the intents/slots or NOT.
This is high priority request from amazon developer and i have seen many developers asking for the same request on amazon forums as well as stackoverflow, etc
One thing to note is that, even Google assistant and other platforms like Microsoft cortana are also providing the whole user utterance whether they recognize intent/slots or not.
And i, as a developer expect the same behavior from Amazon alexa as well and proving the full user utterance, a developer can serve the user in a lot better ways.