Preloading contextual information for applications using a conversation assistant

摘要

Usage data associated with a user of a telephonic device is accessed by a remote learning engine. A service or a product is identified by the remote learning engine based on the accessed usage data. A recommended voice bundle application is determined by the remote learning engine. A recommendation associated with the recommended voice bundle application is transmitted to the telephonic device. The recommendation is presented to the user through voice communications. One or more input parameters associated with the recommended voice bundle application is collected by the telephonic device. The user through voice communications has accepted the recommendation determining is determined. In response to determining that the user has accepted the recommendation, the one or more input parameters to the recommended voice bundle application are loaded by the telephonic device, and the recommended voice bundle application is executed by the telephonic device.

主权项

1. A computer-implemented method comprising:
accessing, by a learning engine, usage data of a telephonic device; examining, by the learning engine, the usage data; determining, by the learning engine based on examining the usage data, a first voice bundle for recommending to a user of the telephonic device, the first voice bundle including first instructions that, when executed by the telephonic device, results in a simulated multi-step spoken conversation between the telephonic device and the user to enable the user to receive a first service or a first product; providing, by the learning engine to the telephonic device, first information for recommending the first voice bundle to the user; in response to providing the information to the telephonic device, obtaining, at the learning engine, feedback provided by the user; examining, by the learning engine, one or more of the feedback and the usage data; determining, by the learning engine and based on examining one or more of the feedback and the usage data, whether a second voice bundle is available for recommending to the user, the second voice bundle including second instructions that, when executed by the telephonic device, results in a simulated multi-step spoken conversation between the telephonic device and the user to enable the user to receive a second service or a second product, the second voice bundle being different from the first voice bundle; and conditioned on determining that the second voice bundle is available, providing, by the learning engine to the telephonic device, second information for recommending the second voice bundle to the user.