Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
METHOD FOR CONTROLLING VOICE RECOGNITION MODULE, VOICE RECOGNITION MODULE, AND HOUSEHOLD APPLIANCE
Document Type and Number:
WIPO Patent Application WO/2021/001255
Kind Code:
A1
Abstract:
A method for controlling a voice recognition module of a household appliance, a voice recognition module of said household appliance, and a household appliance are provided. The voice recognition module is in a dormant state, and the voice recognition module includes a microphone array and a speaker. Steps of the method include: at least one microphone being in an awake state, and controlling the speaker to send a sound wave signal after the at least one microphone receives a sound; the at least one microphone receiving an echo of the sound wave signal, and obtaining a current echo graph in a space where the voice recognition module is located; and comparing the current echo graph with a preset echo graph, so as to control the voice recognition module. By means of this solution, the voice recognition module can be woken up when it is detected that a user needs to use a voice recognition function, thereby reducing energy consumption and costs.

Inventors:
WU TIANLU (CN)
YANG JIA (CN)
LEI DONGYONG (CN)
Application Number:
EP2020/067855
Publication Date:
January 07, 2021
Filing Date:
June 25, 2020
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
BSH HAUSGERAETE GMBH (DE)
International Classes:
G10L15/22; G01S15/52
Foreign References:
US20160044394A12016-02-11
Download PDF:
Claims:
CLAIMS

What is claimed is:

1. A method for controlling a voice recognition module of a household appliance, wherein the voice recognition module is in a dormant state, and the voice recognition module comprises a microphone array and a speaker, characterized by comprising: at least one microphone being in an awake state, and controlling the speaker to send a sound wave signal after the at least one microphone receives a sound; the at least one microphone receiving an echo of the sound wave signal, and obtaining a current echo graph in a space where the voice recognition module is located; and comparing the current echo graph with a preset echo graph, so as to control the voice recognition module.

2. The method according to claim 1 , characterized by comprising, before the method, obtaining the preset echo graph when the space where the voice recognition module is located is an empty room.

3. The method according to at least one of the preceding claims, characterized by further comprising: controlling the entire voice recognition module to wake up when the current echo graph is different from the preset echo graph.

4. The method according to at least one of the preceding claims, characterized by further comprising: controlling the entire voice recognition module to wake up when a difference between the current echo graph and the preset echo graph exceeds a preset value.

5. A voice recognition module of a household appliance, comprising a microphone array, a speaker, and a controller, characterized in that

the controller is configured to keep at least one microphone in an awake state, enable the speaker to send a sound wave pulse signal after the at least one microphone receives a sound, and enable the at least one microphone to receive the sound wave pulse signal; and is further configured to obtain a current echo graph in a space where the voice recognition module is located, and compare the current echo graph with a preset echo graph, so as to control the voice recognition module.

6. The voice recognition module according to claim 5, characterized in that the preset echo graph is obtained when the space is an empty room.

7. The voice recognition module according to at least one of claims 1 to 6 , characterized in that the controller is further configured to wake up the entire voice recognition module when the current echo graph is different from the preset echo graph.

8. The voice recognition module according to claim 5 or 7, characterized in that the controller is further configured to wake up the entire voice recognition module when a difference between the current echo graph and the preset echo graph exceeds a preset value.

9. A household appliance, comprising the voice recognition module according to any one of claims 5 to 8.

Description:
METHOD FOR CONTROLLING VOICE RECOGNITION MODULE, VOICE RECOGNITION MODULE, AND HOUSEHOLD APPLIANCE

BACKGROUND

Technical Field

The present invention relates to the field of household appliance technologies, and in particular, to a method for controlling a voice recognition module, a voice recognition module, and a household appliance.

Related Art

With the development of smart household appliance technologies, voice recognition control technologies are increasingly applied to household appliances. Generally, a voice module of a household appliance has two modes: an operating mode and a standby mode. In an existing method, when the voice module is in the standby mode, a sensor is used to activate or wake up the voice module, so that the voice module enters the operating mode. However, such a method requires the configuration of a sensor in the household appliance, increasing costs of the household appliance.

SUMMARY

An objective of embodiments of the present invention is to wake up a voice recognition module by using a microphone array and a speaker of the voice recognition module without additionally disposing any sensor, thereby simplifying the design of structure lines and reducing equipment costs.

To solve the foregoing problem, the present invention provides a method for controlling a voice recognition module, where the voice recognition module is in a dormant state, and the voice recognition module includes a microphone array and a speaker, at least one microphone is in an awake state, and the speaker is controlled to send a sound wave signal after the at least one microphone receives a sound; the at least one microphone receives an echo of the sound wave signal, and obtains a current echo graph in a space where the voice recognition module is located; and the current echo graph is compared with a preset echo graph, so as to control the voice recognition module. By means of this solution, the voice recognition module can be woken up when it is detected that a user needs to use a voice recognition function, thereby reducing energy consumption and costs.

In a possible embodiment, before the method, the preset echo graph is obtained when the space where the voice recognition module is located is an empty room. Whether there is a user entering the space can be recognized by comparing a different echo graph formed in a different space from a sound wave with a graph formed in the empty room.

In a possible embodiment, the method further includes controlling the entire voice recognition module to wake up when the current echo graph is different from the preset echo graph.

In a possible embodiment, the method further includes controlling the entire voice recognition module to wake up when a difference between the current echo graph and the preset echo graph exceeds a preset value.

The embodiments of the present invention further disclose a voice recognition module including a microphone array, a speaker and a controller, where the controller is configured to keep at least one microphone in an awake state, enable the speaker to send a sound wave pulse signal after the at least one microphone receives a sound, and enable the at least one microphone to receive the sound wave pulse signal; and is further configured to obtain a current echo graph in a space where the voice recognition module is located, and compare the current echo graph with a preset echo graph, so as to control the voice recognition module.

In a possible embodiment, the preset echo graph is obtained when the space is an empty room.

In a possible embodiment, the controller is further configured to wake up the entire voice recognition module when the current echo graph is different from the preset echo graph. In a possible embodiment, the controller is further configured to wake up the entire voice recognition module when a difference between the current echo graph and the preset echo graph exceeds a preset value.

A household appliance includes the voice recognition module according to any one of the foregoing.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a schematic structural diagram of a voice recognition module according to an embodiment of the present invention; and

FIG. 2 is a flowchart of a method for controlling a voice recognition module according to an embodiment of the present invention.

DETAILED DESCRIPTION

As shown in FIG. 1 , a voice recognition module 100 includes a controller 101 , a microphone array 102 connected to the controller 101 , and a speaker 103, and further includes a power supply unit 104 supplying power to the controller 101 , the microphone array 102, and the speaker 103.

Generally, an operating mode of the voice recognition module 100 includes a standby mode and a wakeup mode. It is beneficial to reduction of energy consumption to keep the voice recognition module in the standby mode without using a voice recognition function. When the voice recognition function needs to be used, the voice recognition module 100 is woken up.

The controller 101 may be configured to control operating states of the microphone array 102 and the speaker 103. For example, in an embodiment, the microphone array 102 may include two or more microphones. The controller 101 may be configured to keep at least one microphone in an awake state, and wake up the remaining microphones to operate according to different operating circumstance requirements. An arrangement of the microphones in the microphone array 102 is not limited, may be a linear array, or may be a two-dimensional array or a three-dimensional array.

Specifically, in an embodiment, the voice recognition module 100 may include four microphones and one speaker 103. In a standby mode, the controller 101 is configured to control one of the microphones to be in an awake state. When the microphone receives any sound, the controller 101 controls the speaker 103 to send a sound wave pulse signal. The microphone receives an echo of the sound wave pulse signal, and moreover the controller 102 stores an echo graph and compares the echo graph with a preset echo graph. The controller 101 controls an operating state of the voice recognition module according to a comparison result.

If the comparison result is that the current echo graph is the same as the preset echo graph, it indicates that it is unnecessary to wake up the voice recognition module 100; if the comparison result is that the two are different, it indicates that there is a user entering a current space where the voice recognition module 100 is located and the voice recognition function may need to be used, and the controller 101 performs control to start the remaining three microphones.

In an embodiment, a preset echo graph may be an echo graph of an empty room obtained through a test or an experiment. In this embodiment, when the current echo graph is different from the echo graph of the empty room, it is considered that there is a user entering the same space and the voice recognition function needs to be enabled, and the controller 101 performs control to start corresponding microphones.

In an embodiment, a microphone may be controlled to operate by setting a threshold of a difference between a current echo graph and an echo graph of an empty room, that is, when the difference exceeds the threshold, the remaining microphones are started.

By means of the foregoing solution, when a user needs to use the voice recognition module, the module is woken up, thereby reducing energy consumption and costs. As shown in FIG. 2, a method for controlling a voice recognition module according to an embodiment of the present invention is applicable to the foregoing voice recognition module 100. Steps of the method include the following:

S100, at least one microphone being in an awake state, after the at least one microphone receives a sound, a speaker is controlled to send a sound wave signal.

In an embodiment, one microphone in a microphone array may be in an awake state, and is configured to activate the speaker through any sound received.

In an embodiment, the speaker may send a plurality of sound wave pulse signals to form a clear and accurate sound wave echo graph.

S102, the at least one microphone receives an echo of the sound wave signal and obtains a current echo graph in a space where the voice recognition module is located.

S104, the current echo graph is compared with a preset echo graph to control the voice recognition module.

In an embodiment, when the current echo graph is different from the preset echo graph, it indicates that there may be a user entering the space where the voice recognition module is located and the entire voice recognition module is controlled to wake up, that is, the remaining microphones in a dormant state.

In an embodiment, when a difference between the current echo graph and the preset echo graph exceeds a preset value, the entire voice recognition module is controlled to wake up.

When the space where the voice recognition module is located is an empty room, the preset echo graph in a state of the empty room is obtained.

Embodiments of the present invention further disclose a household appliance including the foregoing voice recognition module 100. The household appliance may be a refrigerator, a washing machine, a cooker hood, a cooktop, a coffee machine, a dishwasher, or the like.

A refrigerator is taken as an example below, the refrigerator is placed in a room 1 , and the refrigerator is provided with the voice recognition module 100 through which a voice recognition function and a voice control function can be realized. When there is no one in the room 1 , an echo graph X of an empty room of the room 1 is obtained and stored.

Generally, the voice recognition module 100 of the refrigerator is in a standby state to save electric energy, where at least one microphone is left in an operating state. When the microphone 102 receives any sound, the speaker 103 is controlled to send a sound wave pulse signal, the microphone 102 receives an echo of the pulse signal and obtains an echo graph Y, and the echo graph Y is compared with a preset echo graph X of the empty room. If Y is different from X, it indicates that there is someone entering the room 1 and the voice recognition function needs to be enabled, and the controller 101 wakes up the remaining microphones. Otherwise, the voice recognition module 100 continues to be on standby.

A person of ordinary skill in the art may understand that all or part of the steps of all methods of the foregoing embodiments may be completed by a program instructing related hardware, the program may be stored in a computer-readable storage medium, and the storage medium may include: a ROM, a RAM, a magnetic disk, an optical disc, or the like.

Although the present invention is disclosed above, the present invention is not limited thereto. Any person skilled in the art can make various changes and modifications without departing from the spirit and the scope of the present invention, and therefore the protection scope of the present invention should be subject to the scope defined by the claims.