Browse Prior Art Database

Urgency-Based Optimization for Intelligent Personal Assistants

IP.com Disclosure Number: IPCOM000250398D
Publication Date: 2017-Jul-11
Document File: 3 page(s) / 160K

Publishing Venue

The IP.com Prior Art Database

Abstract

Disclosed is a system that optimizes the urgent response from intelligent personal assistant by analyzing both the characteristics of the user’s voice when verbally making an inquiry and the content of the query. The system returns the most relevant answers in an appropriate format/medium.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 52% of the total text.

1

Urgency-Based Optimization for Intelligent Personal Assistants

When intelligent personal assistants process queries, the system returns popular items filtered and executed no matter the conditions or immediate need. The existing technology in this domain does not consider the urgency, severity, or special needs for the user request. A method or system is needed that can optimize intelligent personal assistants' response based on urgency.

The novel contribution is a system that optimizes the urgent response from intelligent personal assistants. The system detects a conversation with a personal cognitive assistant and analyzes the conversation for severity and urgency to optimize the response and subsequent action based on the requests’ severity and urgency. The system might compare intensity to a user baseline or a preset baseline.

In detailed implementation steps, the system:

1. Integrates with the personal cognitive assistant agent 2. Analyzes the speech:

A. Pace B. Tone C. Volume

3. Compares the current voice with this user's previous voice profile to identify the delta between the current voice and this user's regular voice in the pitch tone and volume

4. Analyzes the information provided by the user after the speech is converted to the text

5. Evaluates the severity of the user's condition by calculating the results from the two previous steps. If the system identifies dramatic changes in the voice's pace, volume, and steadiness (i.e., jittering, etc.) then the system identifies urgency and increases the rank of the request.

6. Stores the calculated result in the voice score variable 7. Determines the user’s needs based on the user's description of the speech

content 8. Searches the central repository to find the top ranked choices for the user 9. Maps the search results based the urgency and severity of the user's needs 10.Displays the search results to the user based on the location, quality,

time/schedule, availability of the service 11.Automatically plays the first result, or routes to the closest result based on

severity

The user can then decide to act (e.g., go to an emergency center, etc.) based on the list of suggestions. This system sorts the list based on the urgency, rather than branching or changing the state in a workflow.

2

Figure: Process flow

Example Embodiment #1: Medical Scenario

User A took a new medication and experienced an allergic reaction in the form of troubled breathing. User A activates the personal cognitive assistant agent and shouts, "I just took medication Z. I am having a hard time breathing. I am shaking. I can't hold onto the phone very tight. Please help me. Please find me the nearest h...