ASR is supported byCMU Sphinx andAmazon Lex. All recognition is offline by default.

Whether Lex manages to resolve and fulfill an intent or not, it will still return the recognised utterance in text (unless it did not hear anything at all).LexASR andCMUSphinxASRprovide a methodwaitForUtterance() which blocks until the speech to text service returns a string.

If Lexdoes manage to resolve and fulfill (to the point where it delegates to client-side fulfillment) an intent byinvoking a Lamba function thenLexRecognizernotifies aNlpResultListener that the request has been fulfilled or failed etc.

NlpProviderdefines a methodprocessUtterance() which takes a string utterance and context.LexNlpimplementsNlpProvider and notifies theNlpResultListener.

TTS is supported byMaryTTS andAmazon Polly. Speech synthesis is offline by default.