Zammo uses a default setting for the time after the user speaks and the bot responds to ensure the whole prompt is heard. For example, when the user is asking a question or answering an open-ended question, ample time might be needed for the user to pause, think, and speak. However, there may be situations where a shorter delay is preferable. For example, if the user is answering a simple yes or no question, less time is needed to speak.

One way to reduce the delay is to use neural voices, provided by Microsoft ACS and built right into the Zammo product. For instructions on how to use neural voices, please see the article: Personalize: Neural Voice.

(Fun fact: users can press the pound (#) key after speaking to indicate the end of their turn.)

The delay time can be set within the bot design for an entire conversation or for certain parts of the conversation using the custom event zammoSetTelephonyConfiguration. Instructions on how to implement this custom event can be found in this article: Leveraging Speech Controls in UI Builder.