AUTOMATED ASSISTANT INVOCATION OF APPROPRIATE AGENT

摘要

Invoking an agent during a conversation between a user and an automated assistant. Some implementations are directed to receiving natural language input from a user, during a person-to-automated assistant conversation, indicating a desire to associate an agent, but not indicating a specific agent to be associated with. These implementations relate to selecting a particular agent from a plurality of available agents and forwarding a call request to the selected particular agent. In some implementations, an agent selection model, such as a machine learning model, may be utilized to select the particular agent. The machine learning model generates, for each of a plurality of available agents (and optionally intent(s) for those agents), an output indicating the probability that the available agent (and optionally intent) will produce appropriate responsive content. can be trained to generate