Method and device for recognizing speech input and household appliance

Abstract

A method for detecting a voice input (107) of a user (100) for a household appliance (105) comprises detecting the distance of the user (100) in relation to the household appliance (105), determining the user's (100) intention to use the appliance (105) using the distance, and outputting a signal that indicates to the user (100) that the household appliance (105) is able to actively respond to the voice input (107) if the intention to use the appliance (107) was determined in the determination step.

Description

[0001] The invention relates to a method and a device for recognizing speech input and a household appliance.

[0002] DE 10 2020 208 345 A1 discloses a method for controlling a household appliance by a user.

[0003] The approach presented here aims to create an improved method and device for recognizing speech input, as well as an improved household appliance.

[0004] According to the invention, this problem is solved by a method and a device for recognizing speech input and a household appliance with the features of the main claims. Advantageous embodiments and further developments of the invention are described in the dependent claims.

[0005] The advantage achievable with the invention consists of the possibility of using simplified speech input by recognizing the user and their intention.

[0006] The invention comprises a method for recognizing a user's voice input for a household appliance, wherein the method comprises the following steps: Determining the distance of the user in relation to the household appliance; Determining a user's intent to use the distance; and Emitting a readiness signal that indicates to the user that the household appliance is able to actively respond to an input when the intention to use has been determined in the detection step.

[0007] According to one embodiment, the household appliance can be an oven. According to an alternative embodiment, however, the household appliance can also be a cleaning appliance, a refrigerator, or a beverage vending machine. Advantageously, the described approach can be used to improve many different household appliances. By detecting the intended use and simultaneously signaling to the user that the household appliance is ready to receive voice input, shorter voice commands can be used to control the appliance instead of unnatural, complicated commands. This is made possible because both the household appliance and the user can recognize in advance that they want to communicate with each other. The distance can, for example, be a gap in centimeters or a few meters between the user and the household appliance.If the user is relatively close to the appliance, it can be assumed that they intend to use it. The appliance can indicate this readiness by emitting a ready signal. This intention to use the appliance could, for example, be to start it up or open a door.

[0008] In the detection step, the distance can be determined using at least one sensor. Known sensors can be used for this purpose. According to one embodiment, the sensor system can be a combination of sensors to determine the user's distance more accurately.

[0009] In the detection step, the intention to use the appliance can be determined if the distance is less than a threshold. According to one embodiment, this threshold can represent a value below which, if the distance between the appliance and the user is less than the threshold, the sensor registers the user's approach as an intention to interact. Optionally, the threshold can be adjustable. This allows, for example, spatial conditions or user habits to be taken into account.

[0010] During the output step, the readiness signal can be displayed visually and, additionally or alternatively, audibly. This allows a user to easily indicate that the appliance can recognize voice input. Thus, in a situation where multiple appliances can respond to voice input, it can be displayed which appliance is currently responding. If it is the correct appliance, the user can use a very short voice command, as they do not need to address the appliance separately.

[0011] The procedure can optionally include a step of recording the voice input to request a function of the household appliance. According to one embodiment, the recording step can be triggered by the detection of the intended use. Advantageously, recording can thus only be enabled once the intended use has been detected. For example, the voice input can be recorded using a microphone.

[0012] Alternatively, the process can include a step of executing the function that responds to voice input. This allows for simpler operation. For example, an oven door can be opened by voice command if the user is holding something in their hands.

[0013] A device for recognizing a user's voice input can be trained to execute and / or control the steps of the procedure in appropriate units.

[0014] The approach presented here thus also creates a device designed to carry out, control, and implement the steps of a variant of the method presented here in appropriate facilities. This embodiment of the invention in the form of a device also allows the problem underlying the invention to be solved quickly and efficiently.

[0015] The device can be configured to read input signals and, using these input signals, determine and provide output signals. An input signal can, for example, be a sensor signal readable via an input interface of the device. An output signal can be a control signal or a data signal that can be provided at an output interface of the device. The device can be configured to determine the output signals using a processing instruction implemented in hardware or software. For example, the device can include a logic circuit, an integrated circuit, or a software module and may be implemented as, or comprised of, a discrete component.

[0016] The household appliance may include a device for recognizing voice input. Advantageously, the device may be integrated into the household appliance. Alternatively, at least some of the device's components may be housed in a separate unit located / to be located separately from the household appliance.

[0017] A computer program product may include program code for carrying out the procedure if the computer program product is executed on a device.

[0018] A computer program product or computer program with program code that can be stored on a machine-readable medium such as semiconductor memory, hard disk memory, or optical memory is also advantageous. If the program product or program is executed on a computer or device, it can be used to carry out, implement, and / or control the steps of the method according to one of the embodiments described herein.

[0019] Although the described approach is based on a household appliance, the approach described here can be used accordingly in connection with a commercial or professional device.

[0020] An embodiment of the invention is shown purely schematically in the drawings and is described in more detail below. It shows Fig. 1 a schematic representation of a scenario for an exemplary implementation for recognizing a speech input; Fig. 2 a schematic representation of an embodiment of a household appliance with a device for recognizing speech input; and Fig. 3 an exemplary embodiment of a flowchart of a method for recognizing speech input.

[0021] Fig. Figure 1 shows a schematic representation of a kitchen with a user 100. It depicts a scenario of an exemplary implementation in which the user 100 intends to use a household appliance 105 located in the kitchen, for example, an oven. The user 100 gives a voice command 107: "Open door!", while standing in a limited area 110 in front of the household appliance 105.

[0022] The household appliance 105 is designed to detect the distance of the user 100 in relation to the household appliance 105. Fig. For example, it is recorded that user 100 is located within area 100, which, for instance, covers the immediate vicinity of household appliance 105. This indicates that user 100 intends to use the household appliance 105. The household appliance 105 is designed to determine user 100's intention to use it based on the small distance between the appliance and user 100. Once the household appliance 105 has determined the intention to use it, it is designed to emit a ready signal indicating to user 100 that the appliance 105 is now ready to actively respond to a voice command, in this case, the voice input 107: "Open the door!", and in this case, to open the door of the household appliance 105.

[0023] According to one embodiment, the household appliance 105 has a device for detecting voice input 107. According to one embodiment, the device includes sensors configured to detect the distance of the user 100 to the oven using at least one known sensor type, such as UWB (Ultra-Wideband), radar, LiDAR, or TOF sensors. According to one embodiment, the distance of the user 100 is measured in centimeters or a few meters, with the sensors detecting the designated area 110, for example, within a radius of a maximum of 1 meter. According to an alternative embodiment, the radius can be a maximum of 2 meters. According to one embodiment, the area is limited to a predetermined opening angle to ensure that the user 100 is located directly in front of the household appliance 105.

[0024] Based on the known distance between user 100 and the oven, the device can now determine whether user 100 intends to use the oven. According to this embodiment, user 100 intends to use the oven because they are standing very close to it, within the marked area 110.

[0025] Responding to the detection of the user's intention, the device can output a readiness signal which, according to one embodiment, signals to the user 100 either acoustically and visually, or only acoustically or only visually, that the oven is now able to actively respond to the voice input 107.

[0026] This means that voice commands for controlling the Miele appliance 105 don't need to be very precise. For example, it's not necessary for the user to give 100 complex commands like, "Alexa, tell Miele to open the oven door." This input method is very unnatural for humans. It's better to use a single command for voice input that more closely resembles natural human communication, such as the voice command 107: "Open the door." This allows for a convenient shortening of voice commands for triggering functions on household appliances like the Miele appliance 105.

[0027] According to one embodiment, it is possible to uniquely assign the command, here the voice input 107, to a specific device, here the household appliance 105, because the sensor detects where the user 100 is located. Advantageously, the command can also be uniquely assigned to the household appliance 105 in the scenario shown, where optionally several devices in a room have a voice input mode. For example, the approach described here can be used accordingly for a stove 120 or a microwave oven 125.

[0028] According to one embodiment, the household appliance 105 is designed to indicate activation of a voice control by the ready signal as so-called user feedback, for example by a microphone symbol backlighting on a control panel of the household appliance 105.

[0029] According to one embodiment, the household appliance 105 has a door opening function via voice control, which enables voice-controlled oven door opening.

[0030] To detect the distance of the user 100, human presence detection systems in different technological forms (IR reflective light barriers, PIR, RADAR, LiDAR, TOF sensors and cameras, as well as UWB communication) and levels of complexity can be used.

[0031] The readiness signal can be given as feedback to the user of a human-machine interface in various forms, for example symbol illumination, a static or animated display on a matrix display, acoustic feedback such as tones, sounds or speech output.

[0032] According to one embodiment, the proximity of the user 100, for example within an interaction zone, is determined by means of suitable sensors or a combination thereof (RADAR, LiDAR, TOF sensors and camera, as well as UWB wearables). Fig. As shown in Figure 2, the household appliance 105 has been identified. It can now be concluded with a high degree of probability that user 100 has an interaction intention, also referred to as usage intention. In order to indicate to user 100 that the household appliance 105 in their vicinity (interaction zone) is now actively responding to their voice input 107, it is advisable to communicate this status via an appropriate feedback channel, for example, visually via symbol illumination, static or animated displays on matrix displays, or acoustically, such as tones, sounds, or speech output.

[0033] According to one embodiment, human presence detection (HPD) is used as a means of shortening voice commands.

[0034] Fig. Figure 2 shows a schematic representation of an embodiment of a household appliance 105 with a device 200 for carrying out a method for recognizing a speech input 107. For example, the household appliance 105 could be the one described in Fig. The device 200 can be fully or partially integrated into the household appliance 105. Alternatively, the device 200 can also be arranged completely outside of and at a distance from the household appliance 105.

[0035] The device 200 comprises a detection device 210, a locking device 215 and an output device 220. Optionally, the device 200 comprises a receiving device 225 and additionally or alternatively an executing device 230.

[0036] The detection device 210 is configured to detect the distance of the user 100 relative to the household appliance 105 and, for example, to provide a distance-indicating value in the form of a distance signal 240 to the locking device 215. According to one embodiment, the detection device 210 comprises a sensor configured to detect the distance of the user 100 within the characterized area 110. According to one embodiment, the sensor can be a known sensor type, such as IR reflective photoelectric sensors, PIR, radar, LiDAR, time-of-flight sensors and cameras, as well as UWB communication. For example, the area 110 is divided into an approach zone and an interaction zone, with the interaction zone extending between the household appliance 105 and the approach zone.According to one embodiment, the detection device 210 is configured to recognize when user 100 approaches the household appliance 105 by detecting that user 100 enters the approach zone. However, as long as user 100 is within the approach zone, the distance signal 240 is not yet provided, since it is not yet certain whether user 100 actually intends to use the household appliance 105. According to another embodiment, the distance signal 240 is only provided when the detection device 210 detects user 100's presence within the interaction zone. For example, the interaction zone extends from the front of the household appliance 105 to approximately an arm's length away, and the approach zone extends beyond the interaction zone to the maximum sensor range of a sensor used by the detection device 210 to detect user 100.Depending on the implementation, the distance signal 240 can thus include an actual distance measurement, for example in centimeters, decimeters, or meters, or an indication that the user 100 is within the interaction zone. Optionally, the boundary between the interaction zone and the approach zone can be set using a threshold value.

[0037] The detection device 215 is designed to determine the user's intention to use the appliance by means of the distance signal 240 and, according to one embodiment, to provide this information to the output device 220 in the form of a detection signal 245. The user's intention to use the appliance is determined, for example, only if a result from the detection device 210 indicates that the distance between the user 100 and the household appliance 105 is less than the threshold value, i.e., that the user 100 is, for example, within the interaction zone.

[0038] The output device 220 is configured to output a ready signal 250 in response to the confirmation signal 245, which signals to the user 100 that the household appliance 105 is capable of actively responding to the voice input 107. According to one embodiment, the output device 220 comprises a light source for outputting the ready signal 250 visually or a loudspeaker for outputting the ready signal 250 as sound. The output device 220 is also configured to output a signal 255 to the recording device 225 in response to the confirmation signal 245.

[0039] The recording device 225 is configured to record the user's 100 speech input 107 in response to the output signal 255. Alternatively, the recording device 225 can also initiate speech recognition in response to the detection signal 245. According to one embodiment, the recording device 225 includes a microphone.

[0040] In response to the voice input 107, a request signal 260 is sent from the receiving device 225 to the execution device 230, according to one embodiment. Upon receiving the request signal 260, the execution device 230 complies with the command from the voice input 107, here: "Open door!". According to one embodiment, the execution device 230 comprises a motor or an actuator, which, for example, opens the door of the household appliance 105.

[0041] Fig. Figure 3 shows an exemplary embodiment of a flowchart for a method 300 for recognizing speech input. The method can be implemented, for example, using a device such as those described in Fig. 2 is described.

[0042] In step 301 of the data acquisition process, the distance of the user in relation to the household appliance is recorded, for example using sensors.

[0043] In step 303 of the determination process, the user's intent to use the service is determined based on the recorded distance. In this step, the user's distance is compared to a threshold value. If the distance is less than the threshold, intent to use the service can be determined. Otherwise, intent to use the service is not assumed.

[0044] In step 305 of the output process, a signal is then emitted to determine the user's intent, indicating to the user that the household appliance is capable of actively responding to voice input. This signal is emitted, for example, as a light or a sound.

[0045] Method 300 optionally includes a recording step 307. In recording step 307, the user's voice input is recorded. The user can request a function of the household appliance through this voice input. For example, step 307 is executed responsively to the determination of the user's intent, such as after step 305, which outputs the signal. If voice input can only be recorded after the intent has been determined, it can be ensured that the correct household appliance is controlled by the voice input. According to one embodiment, recording step 307 is performed using a microphone.

[0046] Procedure 300 can optionally include an execution step 309. In execution step 309, the function is executed in response to the voice input, which is recorded in step 307. According to one embodiment, a door of the household appliance is opened upon the command "Open door!" QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] DE 10 2020 208 345 A1

[0002]

Claims

[1] Method (300) for recognizing a voice input (107) of a user (100) for a household appliance (105), wherein the method (300) comprises the following steps: Determining (301) a distance of the user (100) in relation to the household appliance (105); Determining (303) a user's intention to use (100) by means of distance; and Output (305) of a ready signal (250) that signals to the user (100) that the household appliance (105) is able to actively respond to the voice input (107) when the intention to use has been determined in the detection step (303). [2] Method (300) according to claim 1, wherein in the step (301) of detection the distance is detected using at least one sensor. [3] Method (300) according to one of the preceding claims, wherein in step (303) of determining the intention to use is determined if the distance is less than a threshold value. [4] Method (300) according to one of the preceding claims, wherein in the step (305) of output the readiness signal (250) is output visually and / or acoustically. [5] Method (300) according to one of the preceding claims, comprising a step (307) of recording the voice input (107) to request a function of the household appliance (105), responding to the determination of the intention to use. [6] Method (300) according to claim 5, wherein in the step (307) of recording the speech input (107) is recorded using a microphone. [7] Method (300) according to claim 5 or 6, comprising a step (309) of performing the function responding to the speech input (107). [8] Device (200) for recognizing a voice input (107) of a user (100), wherein the device (200) is configured to perform and / or control the steps of the method (300) according to claims 1 to 7 in corresponding units. [9] Household appliance (105) with a device (200) according to claim 8. [10] Computer program product with program code for carrying out the method (300) according to any one of claims 1 to 7, when the computer program product is executed on a device (200) according to claim 8.

Images