A Voice First Glossary

acoustic model
A representation that maps the relationship between a digital signal to that of human spoken audio and the phonemes (see term) and other markers that describe various linguistic characteristics of that audio. Models are built based on sets of audio recordings and human transcription of such recordings.
activation phrase
What the user says to activate a voice first device, for instance, “Hey Google” to wake up a Google smart speaker or a smartphone.
adaptive system
A system that adapts its behavior to changing parameters, such as the user’s identity, time of day, day of the week or month, context of the interaction, etc.
adjacency pair
A term used in linguistics to refer to two utterances by two speakers during a conversation, one after the other. For example, “Are you ready to start the survey?” followed by “Yes!” or “Would you like to subscribe to the service?” followed by “Nope!” are adjacency pairs.
A term that refers to an action potential that presents itself to a user of an object as a result of the object’s shape or form or some other physical property that the user perceives as enabling them to do something with the object. A button invites a user to push it, while a stairway handle invites the user to hold it while climbing up or down the stairs.
always-listening device
A device that is always listening for a wake word and that sends the audio captured after the wake word has been detected for additional processing.
anaphora ...

Get The Elements of Voice First Style now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.