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

A smart speaker wake-up method and terminal based on face recognition

A smart speaker and face recognition technology, applied in character and pattern recognition, energy consumption reduction, data processing input/output process, etc., can solve problems such as consuming user waiting time, poor user experience, and low interaction efficiency, and achieve Improve interaction efficiency and user experience

Active Publication Date: 2022-07-12
FUJIAN LANDI COMML EQUIP CO LTD
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] Whether it is Xiaomi smart speakers, Xperia Hello smart speakers, or other existing smart speakers, the interaction mode is: keyword wake-up-response-input-understanding-feedback, that is to say, users need to interact with smart speakers. The voice command of the wake-up word needs to be issued first. After the user sends the wake-up word command, the user needs to wait for the feedback of the smart speaker before inputting the function command, that is, the user needs to issue the wake-up word and wait for the feedback of the smart speaker before the user can start interacting with the smart speaker. This consumes the user's waiting time, the interaction efficiency is low, and the user experience is poor

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
  • A smart speaker wake-up method and terminal based on face recognition
  • A smart speaker wake-up method and terminal based on face recognition

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0129] Please refer to figure 1 , a smart speaker wake-up method based on face recognition, including:

[0130] S1. Identify the facial features of the user in the current scene;

[0131] Wherein, the S1 includes:

[0132] Identify the facial features of the user in the current scene, the facial features include the number of facial features, the integrity of the nose and the relative size of the eyes, and the facial features are scored according to the number of facial features, the integrity of the nose and the relative size of the eyes;

[0133] Wherein, the scoring of the facial features according to the number of facial features, the integrity of the nose and the relative size of the eyes includes:

[0134] According to the scores of the number of facial features, the integrity of the nose, the relative size of the eyes and their corresponding weight values, the scores of the facial features are obtained by weighting and summing.

[0135] The second image information o...

Embodiment 2

[0145] Please refer to figure 1 , the purpose of this embodiment is to determine the processing method of the interactive user by the smart speaker when there are multiple people facing the smart speaker in the current scene. The difference between this embodiment and the first embodiment is that the S1 includes:

[0146] Obtain the total number of users in the current scene, identify the facial features of all users according to the total number, and score the facial features of all users according to the number of facial features, the integrity of the nose and the relative size of the eyes;

[0147] Wherein, the obtaining the total number of users in the current scene includes:

[0148] receiving the first image information collected by the infrared sensor, and sending a steering instruction to the camera according to the first image information;

[0149] Among them, the infrared sensor can collect images of the current scene in 360° without dead angle. When it senses that ...

Embodiment 3

[0169] Please refer to figure 1 , the purpose of this embodiment is to provide another solution to determine the user interacting with the smart speaker when there are many people in the current scene. The difference between this embodiment and the first embodiment is that the S1 includes:

[0170] S0. Pre-store the facial features of the preset user and set the corresponding interaction priority;

[0171] Specifically, different usage priorities may be divided in advance for the pre-stored faces of each preset user;

[0172] The step S2 also includes:

[0173] Determine whether the number of users with scoring results greater than a preset threshold is greater than 1;

[0174] If so, the facial features of the user greater than the preset threshold are matched with the pre-stored preset facial features of the user. Obtain matching users and their corresponding interaction priorities, and determine users who interact with the smart speaker according to the obtained matching...

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 present invention provides a smart speaker wake-up method and terminal based on face recognition. By identifying the face features of the user in the current scene, it is determined whether there is a user facing the smart speaker according to the face features, and if so, directly wake up The smart speaker is to realize the interaction between the smart speaker and the user facing the smart speaker; it is judged that there is a user facing the smart speaker by identifying the user's facial features to directly wake up the smart speaker for interaction, thereby There is no need for the wake-response steps in the interaction mode of the smart speaker, and the interaction mode is simplified to: input-understanding-feedback, which improves the interaction efficiency and user experience between the user and the smart speaker.

Description

technical field [0001] The invention relates to the field of smart home, in particular to a method and terminal for waking up a smart speaker based on face recognition. Background technique [0002] With the development of science and technology, smart speakers have gradually entered people's lives and brought a lot of convenience to people's lives. Taking the Xiaomi smart speaker on the market as an example, when the user interacts with the Xiaomi smart speaker, they need to issue the wake-up word "Xiao Ai" to wake up. After the Xiaomi smart speaker receives the instruction of the wake-up word, it will give feedback, and then the user can input Function commands for voice communication. Another example is the Xperia Hello, a smart speaker from Sony. Users also need to issue the wake-up word "Hi Xperia" and wait for the Xperia Hello smart speaker to respond before entering functional commands for voice interaction. [0003] Whether it is Xiaomi smart speakers and Xperia He...

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
Patent Type & Authority Patents(China)
IPC IPC(8): G06F3/01G06F3/0484G06V40/16
CPCG06F3/011G06F3/0484G06V40/161G06V40/168Y02P90/02
Inventor 卢智全林魁
Owner FUJIAN LANDI COMML EQUIP CO LTD