Enterprise voice artificial intelligence company Regal Voice Inc. today added Goal Oriented Agents to its platform, enabling businesses using them to define what a call has to accomplish, and providing two AI models that split the work of getting there.
The new agents are aimed at a problem Regal says shows up in calls such as auto insurance claims, in which someone reporting an accident tends to explain what happened, where and who was involved in one go. Agents built on conditional logic often circle back for details the caller already gave, just so each field gets filled. Building the whole agent on a single prompt keeps the conversation open, and in Regal’s view it leaves business processes only loosely followed.
Goal Oriented Agents put two models on each call. Regal calls them an actor and a director, and both start work as soon as the caller stops talking. The actor, a smaller and faster model tuned for dialogue, replies right away. Running behind it, the director checks whether the goals for the current stage have been met and can move the call to a new stage with fresh instructions for the actor.
Businesses build the agents out of goal nodes, each listing the goals to be met by the end of the node and any prerequisites that come first. Booking a callback can count as one of those goals. A prerequisite is something the agent has to establish before it can act, such as the caller’s date of birth. Regal said those can be written as plain-language conditions too, for example “the caller mentioned a death in the family.”
The director reviews progress after every turn, so one answer can tick off several items at once. Regal’s sample auto insurance claim has a caller named Dan describing the vehicle damage, photo evidence and injuries in his first message. All three get marked as collected, and the agent goes straight to his name. Once it has that, the director sees location and policy number are still missing, so the actor’s next question is where he is.
Regal said complex intake and qualification calls are among the strongest uses it has seen for the agents. The company named Medicare enrollment, roadside assistance and mortgage qualification as uses the design opens up. Medical intake that complies with the Health Insurance Portability and Accountability Act is another.
Google LLC, AAA and Toyota Motor Corp. are among the more than 200 enterprises Regal says it serves. Customers of Five9 Inc.’s contact center platform gained access to Regal’s voice agents through an integration in August.
Regal is a venture capital-backed startup that has raised $83 million to date. That includes a $40 million round in October 2024.
Image: Regal