Customer Choice and Agent Invocation

Customer choice is a bedrock principle of the Voice Interoperability Initiative. All available agents a customer wishes to interact with should be available to them simultaneously, and should be invocable and identifiable by their own unique wake words.

Baseline Guidance
  • A customer should be able to choose from available voice agents for a particular interaction. They should have the option to use multiple simultaneous wake words when more than one agent is registered on a device.
  • Multiple simultaneously registered agents should be available to customers at all times, aside from the following exceptions:
    • When one agent has been invoked and is actively streaming a customer utterance to the cloud, no other agent’s wake word should be detectable. For example, if a customer says “Agent 1, tell me about Agent 2,” Agent 2 should not be invoked.
    • An agent should not be able to invoke any other agent by distributing the wake word via TTS. For example, one agent cannot wake up another agent by speaking its wake word.
  • When an agent is in Speaking state, responding to a customer, the customer should be able to interrupt that agent’s response with any other active agent’s wake word (barge in).

Agent Preference

Key to the VII pillar of customer choice is the idea that customers can, at any time, use any of the agents registered on their device. Having a selection of agents to choose from, with different capabilities covering a variety of “hero” use cases, offers an individualized experience which customers can explore in their own ways.


There may also be situations in which customers do not use a specific wake word to start an interaction, for example when pressing an Action button. In those cases, the product may designate, or allow a customer to choose, which agent will respond.


When creating simultaneous multi-agent experiences, it is important to design devices and voice agents which can handle the complexities of invocation, from the variety of wake words to the availability and possible overloading of buttons. Invoking a specific agent should involve simple, easy-to-remember methods.

Wake Words

For many devices, wake words are going to be the primary method customers use to invoke an agent. The wake word, along with the agent’s voice, is also a significant component of persona and brand. If your customers are going to be interacting with your agent using a wake word, keep in mind the following:

  • Wake words should be distinct from each other, and easy to remember, both to avoid customer confusion and to minimize the chance that a device will mistake one wake word for another.
  • If you allow customers to choose alternative wake words to invoke an agent, be aware of other agent wake words and alternative options.