Browse Prior Art Database

Cognitive home security system alarm based on speech, tone and emotions Disclosure Number: IPCOM000250408D
Publication Date: 2017-Jul-11
Document File: 2 page(s) / 56K

Publishing Venue

The Prior Art Database

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 76% of the total text.

Cognitive Home Security System Alarm Based on Speech, Tone, and Emotions Violent crime against occupants of a house/apartment/office can occur without triggering a home security perimeter system. If the security perimeter system is inactivated or disabled, then people within the house are subject to a crime by the invader. Existing solutions require the input of a sensor or a human. Whether there is a panic button or a perimeter detection system, these can be of no use if the criminal has already reached and disabled the victims within premises. The novel contribution is a system for using cognitive processing of speech within a given location to identify potential crime situations. An always-on device that uses a tone analyzer is deployed in an area where it can capture conversations from people on the premises. In a home invasion, domestic violence, or other criminal situation, a dialog presents between the perpetrator and the victim, mostly based on the demands of the perpetrator. This device captures and processes text from the conversations that occur within range and identifies potential crime situations. Using existing Application Programming Interfaces (APIs), the system can also provide critical information (e.g., quantity and estimated ages of the perpetrators) to law enforcement in advance of the team arriving at the crime scene. In addition to the known API tone analyzer, further APIs developed using cognitive solutions for sentiment analysis can be used to bette...