Simulated conversation by pre-recorded audio navigator

摘要

A method is provided for a simulated conversation by a pre-recorded audio navigator, with particular application to informational and entertainment settings. A monitor may utilize a navigation interface to select pre-recorded responses in the voice of a character represented by a performer. The pre-recorded responses may then be queued and sent to a speaker proximate to the performer. By careful organization of an audio database including audio buckets and script-based navigation with shifts for tailoring to specific guest user profiles and environmental contexts, a convincing and dynamic simulated conversation may be carried out while providing the monitor with a user-friendly navigation interface. Thus, highly specialized training is not necessary and flexible scaling to large-scale deployments is readily supported.

主权项

1. A computing device for providing a simulated conversation with a guest using a pre-recorded audio, the computing device comprising:
a processor configured to:
initialize a present state according to a script in a database;receive a navigation input from an operator monitoring the guest, where the navigation input is in response to an action by the guest;modify the present state using the script and the navigation input;determine an audio segment of the pre-recorded audio to add to a queue based on the present state;process the queue to output the audio segment to a speaker proximate to a performer for the simulated conversation.