A computer-readable storage device comprising executable instructions that, when executed by a processor, assist with personalized interaction with a digital assistant, the computer-readable medium including instructions executable by the processor for:
- receiving a first voice input from a first user;
- analyzing acoustic patterns of the first voice input;
- determining, based on the acoustic analysis, the identity of the first user from among a plurality of user identities including the first user and a second user;
- extracting, from the first voice input, at least one signal pertaining to a background characteristic of the first user and at least one signal pertaining to a digital characteristic associated with the first user, the digital characteristic including at least one of a request and keywords;
- assigning values to the at least one extracted signal pertaining to the background characteristic of the first user and the at least one extracted signal pertaining to the digital characteristic associated with the first user;
- categorizing the first voice input into one or more topical categories using the assigned values and performing natural language speech processing on the first voice input using a natural language processing model selected in accordance with the one or more topical categories;
- analyzing world knowledge related to the categorized first voice input from the second user;
- responsive to the at least one signal pertaining to a background characteristic of the first user extracted from the first voice input, the at least one signal pertaining to a digital characteristic associated with the first user extracted from the first voice input, and the analyzed world knowledge related to the categorized first voice input, providing a filtered and proactive response related to the first voice input;
- receiving a second voice input from the second user;
- analyzing, in a second acoustic analysis, acoustic patterns of the second voice input;
- determining, based on the second acoustic analysis, the identity of the second user from among the plurality of user identities;
- extracting, from the second voice input, at least one signal pertaining to a background characteristic of the second user and at least one signal pertaining to a digital characteristic associated with the second user, the digital characteristic including at least one of a request and keywords;
- assigning values to the at least one extracted signal pertaining to the background characteristic of the second user and the at least one extracted signal pertaining to the digital characteristic associated with the second user;
- categorizing the second voice input into the one or more topical categories using the assigned values and performing natural language speech processing on the second voice input using a natural language processing model selected in accordance with the one or more topical categories, wherein keywords determined from the first voice input and the second voice input are the same;
- and
- analyzing world knowledge related to the categorized second voice input from the second user;
- responsive to the at least one signal pertaining to a background characteristic of the second user extracted from the second voice input, the at least one signal pertaining to a digital characteristic associated with the second user extracted from the second voice input, and the analyzed world knowledge related to the categorized second voice input, providing a proactive response related to the second voice input that is different from the filtered response related to the first voice input.