There are more and more voice-enabled products on the market, allowing the user to control them through the voice, without the need of a physical interface. To enhance the interaction experience it is possible to equip a voice-enabled device with a Voice Assistant (VA) that can answer and engage in conversation with the user. This can be very useful especially when user requests are more elaborate and when it is useful to receive feedback from the device. In this regard, one can choose from a wide variety pre-built Voice Assistants which may offer different benefits.
Depending on the Voice Assistant used and on the type of device, there can be several ways to connect them. Sometimes, for example light bulbs, the pre-built Voice Assistants control their host devices directly. Sometimes, for example in a smart oven, the device hosts two local clients: one for the Voice Assistant and the other for the device itself.
In this last case, once the user’s voice is captured by the device microphones and processed by the Acoustic Front End, the local Voice Assistant client sends the audio containing the command to servers of the VA provider. The VA servers transcribe and analyze it to extract the intent and then send it to a logic block which converts the intent into a request. This request is forwarded to the oven manufacturer servers, which convert it into the final command and send it to the client located on the device. At this point the device performs the action and can provide a feedback that follows the same process in reverse, resulting in an audio response from the device’s VA.
Do you want to know more?
This post is a part of our white paper “The whys and hows of voice enabling your product“.
In this white paper we explain how to choose and integrate a voice assistant into a product in order to make informed choices and fulfill customers’ need.
Use this form to receive it directly in your mailbox!