Unlock instant, AI-driven research and patent intelligence for your innovation.

Method and device for performing monitoring by adopting intelligent loudspeaker box

A technology of smart speakers and monitoring environment, applied in general control system, program control, computer control and other directions, can solve the problems of low monitoring efficiency, can not protect user privacy well, expose images, etc., to achieve the effect of improving monitoring efficiency

Active Publication Date: 2019-02-01
SAMSUNG ELECTRONICS CHINA R&D CENT +1
View PDF23 Cites 8 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] 1. Unable to automatically perceive abnormalities, and the monitoring efficiency is low;
[0005] 2. The indoor monitoring method based on the abnormal perception of the image captured by the camera will expose the image of the target place and cannot protect user privacy well;
[0006] 3. The indoor monitoring method based on abnormal sensor perception requires users to install corresponding sensors

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Method and device for performing monitoring by adopting intelligent loudspeaker box
  • Method and device for performing monitoring by adopting intelligent loudspeaker box
  • Method and device for performing monitoring by adopting intelligent loudspeaker box

Examples

Experimental program
Comparison scheme
Effect test

example 1

[0111] Step 01: Set the monitor mode.

[0112] The user sets the smart speaker to monitor mode, and adds the following environmental sounds to the wake-up sound: rain, faucet running water, door opening, sneezing, coughing and alarm.

[0113] Step 02: The smart speaker starts to collect ambient sound, and extracts the acoustic and orientation features of the ambient sound in real time.

[0114] Acoustic features include: WPD coefficient, pitch subband energy, short-term zero-crossing rate, short-term average energy, real-time amplitude or power, adjacent band feature vector, LPCC and MFCC.

[0115] Azimuthal features include: ITD, IID, ILD, and IPD.

[0116] Step 03: Input the acoustic features and orientation features extracted in Step 02 into the sound recognition model, and the voice recognition model outputs the recognized sound types: rain sound and sound source orientation, and compare the sound source orientation with the pre-recorded indoor areas The location range i...

example 2

[0122] Step 01: Set the monitor mode.

[0123] The user sets the smart speaker to monitor mode, and adds the following environmental sounds to the wake-up sound: rain, faucet running water, door opening, sneezing, coughing and alarm.

[0124] Step 02: The smart speaker starts to collect ambient sound, and extracts the acoustic and orientation features of the ambient sound in real time.

[0125] Step 03: Input the acoustic features and orientation features extracted in Step 02 into the voice recognition model, and the voice recognition model outputs the recognized sound types: door opening sound and sound source orientation, and compare the sound source orientation with the pre-recorded indoor areas The location range is matched to determine that the sound source comes from the entrance door.

[0126] Step 04: According to the preset time range of the user returning home, it is determined that the current time is not the time for the user returning home, then it is determined ...

example 3

[0129] Step 01: Set the monitor mode.

[0130] The user sets the smart speaker to monitor mode, and adds the following environmental sounds to the wake-up sound: rain, faucet running water, door opening, sneezing, coughing and alarm.

[0131] Step 02: The smart speaker starts to collect ambient sound, and extracts the acoustic and orientation features of the ambient sound in real time.

[0132] Step 03: Input the acoustic features and orientation features extracted in Step 02 into the voice recognition model, and the voice type recognition model outputs the recognized voice types: sneezing sound and sound source orientation, and compare the sound source orientation with the pre-recorded indoor areas Match the location range of the sound source to determine the room where the sound source is coming from.

[0133] Step 04: Count the number of sneezing sounds. If it exceeds 5 times, judge whether the current indoor temperature is lower than the preset comfortable temperature. If...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention provides a method and a device for performing monitoring by adopting an intelligent loudspeaker box. The method comprises the following steps that for each type of sound possibly occurring in a monitoring environment, according to all possible positions of a sound source of the sound relative to the intelligent loudspeaker box in the monitoring environment, a sound signal sent by thesound source of the sound in each possible position is collected by the intelligent loudspeaker box; the acoustic characteristics of the sound and the orientation characteristics of the sound sourceare extracted from the sound signals; the acoustic characteristics and the orientation characteristics are input into a training model for training, thereby obtaining a sound recognition model used for recognizing the type of the source and the orientation of the sound source; the sound signals in the monitoring environment are collected in real time through the intelligent loudspeaker box, the acoustic characteristics of the sound and the orientation characteristics of the sound source are extracted from the sound signals in real time, and the extracted acoustic characteristics and orientation characteristics are input into the sound recognition model in real time; and according to the type of the sound and the orientation of the sound source, output by the sound recognition model, whether a monitoring scene is abnormal or not is determined. According to the method and the device, the monitoring efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of monitoring. In particular, it relates to a monitoring method and device using an intelligent speaker. Background technique [0002] The number of smart speakers is increasing year by year, and it is estimated that the overall market for smart speakers will reach 563 billion units in 2018. At present, the main functions of smart speakers are: smart home central control, voice shopping, mobile phone recharge, takeaway, audio and music playback, etc. [0003] At present, indoor monitoring methods mainly rely on cameras or sensors to achieve the purpose of monitoring. The disadvantages of this approach are as follows: [0004] 1. Unable to automatically perceive abnormalities, and the monitoring efficiency is low; [0005] 2. The indoor monitoring method based on the abnormal perception of the image captured by the camera will expose the image of the target place and cannot protect user privacy well; [...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
IPC IPC(8): G05B15/02G05B19/418
CPCG05B15/02G05B19/418G05B2219/2642Y02P90/02
Inventor 王壮杨建军
Owner SAMSUNG ELECTRONICS CHINA R&D CENT