Voice assistance device and method thereof

Abstract

Embodiments of the present disclosure relate to a voice assistant device and a method for controlling a voice assistant device. The voice assistant device includes a receiver configured to receive at least one voice input from a user when operating in a wake-up mode. An intent is associated with the at least one voice input from the at least one user. In addition, a probability of the issuance of a subsequent voice input from the at least one user is determined based on at least one of the intent, historical data, and one or more contextual factors. When the probability is greater than a predetermined threshold, an extended wake-up duration of the voice assistant device is estimated. In addition, the duration of the wake-up mode is extended by the extended wake-up duration to receive a subsequent voice input from the at least one user.

Description

Technical Field

[0001] The present subject matter relates generally to digital assistance systems and, more particularly, but not exclusively, to apparatus and methods for controlling voice-assisted devices. Background Art

[0002] A voice assistant is a digital assistant that uses speech recognition, speech synthesis, and natural language processing (NLP) to provide services through specific applications. A trigger word associated with the voice assistant is used to trigger the voice assistant to switch from sleep mode to wake mode. A user can make a request to the voice assistant by first saying the trigger word and then the request.

[0003] During a conversation with a voice assistant, especially when a conversation is ongoing with the voice assistant, the user may have to repeatedly keep saying the trigger word to keep the voice assistant in wake-up mode. By this method, the context of the conversation can be maintained by the voice assistant, and subsequent requests can be easily interpreted by the voice assistant for accurate responses. In the event that the user forgets to use the trigger word during the conversation, the voice assistant may switch to a sleep mode. In this case, further requests from the user may be missed by the voice assistant. The trigger word may have to be said again to start a new conversation chain, which will be a new context for the voice assistant.

[0004] The information disclosed in this background section of this disclosure is only for enhancement of understanding of the overall background of the invention and should not be construed as an admission or any form of suggestion that the prior art is already known to a person skilled in the art. Summary of the invention

[0005] [Technical issues]

[0006] Some of the voice assistant devices may be configured to operate in tracking mode. In these voice assistant devices, during the wake-up mode and when a request is received from the user, the voice assistant device may be configured to provide a response and switch to the tracking mode. In the tracking mode, the voice assistant device is configured to remain in the wake-up mode for a predetermined duration. The user needs to provide further requests within the scope of the predetermined duration. If no request is received from the user during the tracking mode, the voice assistant device may switch to the sleep mode. Typically, the predetermined duration is static and can be a very short period of time. In the case where the user provides a further request after the predetermined duration, such a request may not be received by the voice assistant device. In addition, there is a static time period of this tracking mode, and the user may feel burdened to prepare further requests to be communicated during the static time period. In some cases, when providing a request, the user may not know whether the voice assistant device is operating in the tracking mode. And the request may have to be communicated well before the end of the predetermined duration.

[0007] [Solution to the problem]

[0008] In an embodiment, the present disclosure relates to a method of a voice assistant device. First, at least one voice input from at least one user is received during a wake-up mode of the voice assistant device, and an intent associated with the at least one voice input is identified. A probability of a subsequent voice input from the at least one user is determined based on at least one of the intent, historical data, and one or more contextual factors. When the probability is greater than a predetermined threshold, an extended wake-up duration of the voice assistant device is estimated. In addition, the duration of the wake-up mode of the voice assistant device is extended by the extended wake-up duration to receive a subsequent voice input from the at least one user.

[0009] In an embodiment, the present disclosure relates to a voice assistant device, which includes a receiver configured to receive at least one voice input from at least one user when operating in a wake-up mode. The voice assistant device also includes a processor and a memory communicatively coupled to the processor. The memory stores instructions executable by the processor, which when executed cause the processor to identify an intent associated with at least one voice input from at least one user through the voice assistant device. In addition, the processor is configured to determine a probability of issuance of a subsequent voice input from at least one user based on at least one of the intent, historical data, and one or more contextual factors. When the probability is greater than a predetermined threshold, an extended wake-up duration of the voice assistant device is estimated. In addition, the duration of the wake-up mode is extended by the extended wake-up duration to receive a subsequent voice input from at least one user.

[0010] In an embodiment, the present disclosure relates to a method for controlling a voice assistant device. First, an intent associated with at least one voice input received by the voice assistant device from at least one user is identified. A probability of an issuance of a subsequent voice input from the at least one user is determined based on at least one of the intent, historical data, and one or more contextual factors. When the probability is greater than a predetermined threshold, an extended wake-up duration of the voice assistant device is estimated. In addition, a duration of a wake-up mode of the voice assistant device is extended by the extended wake-up duration to receive a subsequent voice input from the at least one user.

[0011] In an embodiment, the present disclosure relates to a control device for controlling a voice assistant device. The control device is associated with the voice assistant device and includes a first processor and a first memory communicatively coupled to the first processor. The first memory stores instructions executable by the processor, which, when executed, cause the first processor to control the voice assistant device. First, an intent associated with at least one voice input received by the voice assistant device from at least one user is identified. A probability of issuance of a subsequent voice input from at least one user is determined based on at least one of the intent, historical data, and one or more contextual factors. When the probability is greater than a predetermined threshold, an extended wake-up duration of the voice assistant device is estimated. In addition, a duration of a wake-up mode of the voice assistant device is extended by the extended wake-up duration to receive a subsequent voice input from at least one user.

[0012] In an embodiment, the present disclosure relates to a method for determining a probability of an utterance of a subsequent voice input to a voice assistant device. First, in order to determine, based on the intent of the voice input, one or more keywords are extracted from a plurality of words in at least one voice input received by the voice assistant device. A domain associated with the at least one voice input is identified based on the one or more keywords. In addition, a domain matching score is calculated for the identified domain using a plurality of pre-stored domains associated with at least one user. The probability of an utterance of a subsequent voice input to the voice assistant device is determined based on the domain matching score, historical data, and one or more contextual factors associated with at least one user.

[0013] In an embodiment, the present invention relates to a probability determination device for determining a probability of an issuance of a subsequent voice input to a voice assistant device. The probability determination device is associated with the voice assistant device and includes a second processor and a second memory communicatively coupled to the second processor. The second memory stores instructions executable by the processor, which when executed cause the second processor to determine a probability of an issuance of a subsequent voice input to the voice assistant device. First, in order to determine, based on the intent of the voice input, one or more keywords are extracted from a plurality of words in at least one voice input received by the voice assistant device. A domain associated with the at least one voice input is identified based on the one or more keywords. In addition, a domain matching score is calculated for the identified domain using a plurality of pre-stored domains associated with at least one user. The probability of an issuance of a subsequent voice input to the voice assistant device is determined based on the domain matching score associated with at least one user, historical data, and one or more contextual factors.

[0014] The foregoing summary is illustrative only and is not intended to be limiting in any way. In addition to the illustrative aspects, embodiments, and features described above, further aspects, embodiments, and features will become apparent by reference to the drawings and the following detailed description.

[0015] [Beneficial Effects of the Invention]

[0016] Embodiments of the present disclosure are configured to provide an optimal user experience by dynamically determining an extended duration of a wake-up mode of a voice-assisted device.

[0017] Embodiments of the present disclosure provide accurate probabilities of subsequent speech input being uttered by considering the intent of the speech input, historical data, and contextual factors associated with the user. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] The accompanying drawings, which are included in and constitute a part of the present disclosure, illustrate exemplary embodiments and, together with the description, serve to explain the principles disclosed. In the drawings, the leftmost digit of a reference number identifies the drawing in which the reference number first appears. The same reference numerals are used throughout the drawings to represent similar features and components. Some embodiments of systems and / or methods according to embodiments of the subject matter of the present application are now described by way of example only and with respect to the accompanying drawings, in which:

[0019] Figure 1a to Figure 1c An exemplary environment showing a voice assistance device according to some embodiments of the present disclosure;

[0020] Figure 2 A detailed block diagram showing a control device for controlling a voice assistance device according to some embodiments of the present disclosure;

[0021] Figure 3A detailed block diagram showing a probability determination apparatus for determining the probability of issuing a subsequent voice input to a voice assistant device according to some embodiments of the present disclosure;

[0022] Figure 4a to Figure 4b A schematic diagram showing a method for controlling a voice assistance device according to some embodiments of the present disclosure is shown;

[0023] Figure 5a to Figure 5f illustrates an exemplary embodiment of a voice assistance device according to some embodiments of the present disclosure;

[0024] Figure 6a A flowchart showing a method for controlling a voice assistance device according to some embodiments of the present disclosure;

[0025] Figure 6b A flowchart illustrating a method for determining an utterance probability of a subsequent speech input to a speech assistant device according to some embodiments of the present disclosure; and

[0026] Figure 7 A block diagram is shown of an exemplary computer system for implementing embodiments consistent with the present disclosure.

[0027] It will be understood by those skilled in the art that any block diagram herein represents a conceptual view of an illustrative system that embodies the principles of the subject matter of the present application. Similarly, it will be understood that any flow chart, flow chart, state transition diagram, pseudo code, etc. represent various processes that can be generally represented in a computer-readable medium and executed by a computer or processor, whether or not these computers or processors are explicitly shown. DETAILED DESCRIPTION

[0028] As used herein, the word “exemplary” means “serving as an example, instance, or illustration.” Any embodiment or implementation of the subject matter described herein as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments.

[0029] Although the present disclosure may have various modifications and alternative forms, its specific embodiments have been shown in the accompanying drawings by way of example and will be described in detail below. However, it should be understood that the present disclosure is not intended to limit the present disclosure to the disclosed form, but on the contrary, the present disclosure covers all modifications, equivalents and substitutions that fall within the spirit and scope of the present disclosure.

[0030] The terms "comprises," "comprising," or any other variation thereof, are intended to cover non-exclusive inclusions, such that an arrangement, device, or method that includes a series of components or steps includes not only those components or steps, but may also include other components or steps that are not expressly listed or inherent to such arrangement, device, or method. In other words, one or more elements limited by "comprising" in a system or apparatus does not exclude the presence of other elements or additional elements in the system or method without more limitations.

[0031] The terms "comprises," "comprising," or any other variation thereof, are intended to cover non-exclusive inclusion, such that a setup, device, or method that includes a series of components or steps includes not only those components or steps, but may also include other components or steps that are not explicitly listed or inherent to such setup, device, or method. In other words, one or more elements limited by "comprising" in a system or device do not exclude the presence of other elements or additional elements in the system or method without more limitations.

[0032] In the following detailed description of the embodiments of the present disclosure, reference is made to the accompanying drawings which form a part of the present disclosure, and in which specific embodiments in which the present disclosure may be practiced are shown by way of illustration. These embodiments are described in sufficient detail to enable those skilled in the art to practice the present invention, and it is understood that other embodiments may be utilized and changes may be made without departing from the scope of the present disclosure. Therefore, the following description should not be understood in a limiting sense.

[0033] The present disclosure proposes a method and apparatus for controlling a voice assistant device to provide an optimal user experience. When receiving a voice input from at least one user, the voice assistant device is configured to check the probability of issuing a subsequent voice input. If the probability is greater than a predetermined threshold, the wake-up time of the voice assistant device is extended. The present disclosure also proposes determining an extended wake-up duration of the voice assistant device based on the probability. The wake-up duration is dynamically determined based on the context of the voice input, the probability, and various other factors related to at least one user.

[0034] Figure 1aAn exemplary environment of a voice assistant device 101 is shown. The voice assistant device 101 may be configured to perform speech recognition, speech synthesis, and natural language processing (NLP) to provide services to at least one user 102. The voice assistant device 101 may implement artificial intelligence (AI) to provide services to at least one user 102. The voice assistant device 101 may be associated with a trigger for starting a service to at least one user 102. When not in use, the voice assistant device 101 is configured to operate in a sleep mode, and when in use, the voice assistant device 101 is configured to operate in a wake-up mode. When a trigger is provided to the voice assistant device 101, the voice assistant device 101 may switch from the sleep mode to the wake-up mode. In an embodiment, the trigger may be a specific word associated with the voice assistant device 101. When at least one user 102 speaks the trigger, the voice assistant device 101 is activated to listen to the request after the trigger. Typically, the request is spoken to the voice assistant device 101 by at least one user 102. The request may be in the form of a voice input received by the voice assistant device 101. In an embodiment, the voice assistant device 101 may be a dedicated hardware component registered with an account of at least one user 102. In an embodiment, the voice assistant device 101 may be embedded in a user device of at least one user 102. For example, the voice assistant device 101 may be embedded in a smartphone, a smartwatch, a tablet, a laptop, etc. of at least one user 102. In an embodiment, the voice assistant device 101 may be embedded in an infotainment system of a vehicle of at least one user 102. In an embodiment, the voice assistant device may be embedded in an apparatus and configured to control the apparatus using voice commands. The apparatus may include, but is not limited to, a refrigerator, a refrigerator, a television, a washing machine, a music player, etc. The voice assistant device 101 may be any device that receives voice input from at least one user 102 and provides services based on the voice input. In an embodiment, the voice assistant device 101 may be connected to other devices or applications to provide services to at least one user 102. For example, if the voice assistant device 101 is embedded in a vehicle, the voice assistant device 101 may be connected to the vehicle's music system, navigation system, lighting system, etc. In the event that at least one user 102 requests the voice assistant device 101 to be used to navigate to a destination, the voice assistant device 101 may communicate with the navigation system to provide a service to the at least one user 102. Consider that the voice assistant device 101 is placed in a home environment. The voice assistant device 101 may be connected to a home appliance in the home environment to control the operation of the home appliance. The home appliance may be a controller based on a request from the at least one user 102. For example, if the at least one user 102 requests to turn on the AC. The voice assistant device 101 may communicate to control the operation of the AC and turn on the AC. Upon further request, the voice assistant device 101 may also be configured to control the temperature of the AC.

[0035] In an embodiment, the voice assistant device 101 may be associated with one or more output devices to provide a response to at least one user 102. In an embodiment, the one or more output devices may include a display unit, an audio output unit, a light emitting diode (LED), etc. The response may be provided via a visual display or an audio output, or may be indicated via an LED. One or more other methods known in the art may be implemented in the voice assistant device 101 for providing a response to at least one user 102.

[0036] In addition, the voice assistant device 101 may include a processor 103, an I / O interface 104, a memory 105, and a receiver 106. In some embodiments, the memory 105 may be communicatively coupled to the processor 103. The memory 105 stores instructions executable by the processor 103, which, when executed, may cause the voice assistant device 101 to provide services to at least one user 102. In an embodiment, the memory 105 may include one or more modules 107 and data 108. The one or more modules 107 may be configured to perform the steps of the present disclosure using the data 108 to provide services to at least one user 102. In an embodiment, each of the one or more modules 107 may be a hardware unit that may be external to the memory 105 and coupled to the voice assistant device 101. The voice assistant device 101 may be implemented in various computing systems, such as laptops, desktop computers, personal computers (PCs), notebooks, smart phones, tablet computers, e-book readers, servers, network servers, and the like.

[0037] In an embodiment, the voice assistant device 101 may be controlled by a control device 109. The control device 109 may be configured to control the operation of the voice assistant device 101 based on at least one voice input received from at least one user 102. In addition, the control device 109 may be configured to identify an intent associated with at least one voice input from at least one user 102. In an embodiment, the intent may convey the context or meaning of at least one voice input received from at least one user 102. In an embodiment, the intent associated with the at least one voice input may be identified by performing natural language understanding (NLU) on the at least one voice input. In an embodiment, the conversational knowledge of previous voice inputs of the voice assistant device 101 may be used to determine the intent of the received at least one voice input.

[0038] In addition, a probability of issuing a subsequent voice input from at least one user 102 is determined for at least one voice input. The probability may be determined based on at least one of an intent, historical data, and one or more contextual factors. The historical data may be associated with at least one user 102. In an embodiment, the historical data may include past requests of at least one user 102 and corresponding responses provided by the voice assistant device 101. In an embodiment, the historical data may include the time of the request, the date of the request, the location of the request, and other data related to the past request from at least one user 102. In an embodiment, one or more contextual factors may include at least one of a user-related factor, a time-related factor data, and an environment-related factor. In an embodiment, the user-related factor may include a daily routine, a location, a preference, etc. associated with at least one user 102. In an embodiment, the time-related data may include the time of the request, the date of the request, whether the request is in the morning, afternoon, or evening, etc. In an embodiment, the environment-related factor includes connection details and status details of a device or application connected to the voice assistant device 101. One or more other factors that may be used to determine the probability of issuing a subsequent voice input may be included in the one or more contextual factors.

[0039] In an embodiment, a probability determination device 110 associated with the control device 109 and the voice assistant device 101 may be configured to determine the probability of the subsequent voice input being issued. In order to determine the probability of the issuance, the probability determination device 110 may be configured to extract one or more keywords from a plurality of words in the at least one voice input. The one or more keywords may be extracted based on the intention of the at least one voice input. In an embodiment, the one or more keywords may be extracted by assigning a weight to each of the plurality of words of the at least one voice input. In addition, keywords from the plurality of words having a weight greater than a predetermined weight value may be identified as the one or more keywords.

[0040] In an embodiment, the probability determination device 110 may be configured to establish dialog knowledge for at least one voice input based on one or more extracted keywords. In an embodiment, the dialog knowledge may be used to identify the intention of subsequent voice input to the voice assistant device 101 .

[0041] When extracting one or more keywords, a domain associated with at least one voice input may be identified based on the one or more keywords. In addition, a domain matching score is calculated for the identified domain using a plurality of pre-stored domains associated with at least one user 102. The probability determination device 110 determines the probability of issuing a subsequent voice input to the voice assistant device 101 based on the domain matching score associated with at least one user 102, one or more contextual factors, and historical data.

[0042] When determining the probability of issuance of the subsequent voice input, the control device 109 may be configured to compare the probability with a predetermined threshold. In an embodiment, the predetermined threshold may be determined by analyzing historical data using a device connected to the voice assistant device 101. When the determined probability is greater than the predetermined threshold, an extended wake-up duration of the voice assistant device 101 may be determined. In addition, the duration of the wake-up mode of the voice assistant device 101 may be extended to the extended wake-up duration to receive a subsequent voice input from at least one user 102. In an embodiment, the extended wake-up duration may be estimated to be proportional to the probability of issuance of the subsequent voice input. For example, if the determined probability is high, the extended wake-up duration may be determined to last for a longer duration. Similarly, when the determined probability is low but equal to or greater than the predetermined threshold, the extended wake-up duration may be determined to last for a shorter duration.

[0043] In an embodiment, when the probability is determined to be less than a predetermined threshold, the control device 109 may configure the voice assistant device 101 to be in the sleep mode until a trigger to operate the voice assistant device 101 in the wake-up mode is detected.

[0044] Additionally, during the extended wake-up duration, if no subsequent voice input is detected, the voice assistant device 101 may be configured to be in a sleep mode.

[0045] In an embodiment, the control device 109 and the probability determination device 110 may be components of the speech assistant device 101, such as Figure 1b The at least one speech input received by the receiver 106 may be used by the control device 109 to control the speech assistant device 101, and may be used by the probability determination device 110 to determine the probability of the subsequent speech input being issued. In an embodiment, the probability determination device 110 may be a component of the control device 109 (not shown). Figure 1b In an embodiment, the control device 109 and the probability determination device 110 may be a dedicated server or a cloud-based server that communicates with the voice assistance device 101, such as Figure 1c . The voice assistant device 101 may communicate with such a server via a communication network (not shown in the figure). The communication network includes, but is not limited to, direct interconnection, peer-to-peer (P2P) networks, local area networks (LANs), wide area networks (WANs), wireless networks (e.g., using wireless application protocols), controller area networks (CANs), the Internet, Wi-Fi, etc. The I / O interface 104 of the voice assistant device 101 may facilitate sending and receiving data. The received data may include at least one voice input, instructions, etc. from the control device 109. The sent data may include at least one voice input input to the control device 109, a response to at least one user 102, etc.

[0046] Figure 2 FIG. 1 is a detailed block diagram of a control device 109 for controlling a voice assistant device 101 according to some embodiments of the present disclosure.

[0047] The control device 109 may include a first processor 201, a first I / O interface 202, and a first memory 203. In some embodiments, the first memory 203 may be communicatively coupled to the first processor 201. The first memory 203 stores instructions executable by the first processor 201, which when executed may cause the control device 109 to control the voice assistant device 101. In an embodiment, the first memory 203 may include one or more first modules 204 and first data 205. The one or more first modules 204 are configured to perform the steps of the present disclosure using the first data 205 to provide control of the voice assistant device 101. In an embodiment, each of the one or more first modules 204 may be a hardware unit that may be external to the first memory 203 and coupled to the control device 109. The control device 109 may be implemented in various computing systems, such as a laptop, a desktop computer, a personal computer (PC), a notebook, a smart phone, a tablet computer, an e-book reader, a server, a network server, etc.

[0048] The first data 205 and the one or more first modules 204 in the first memory 203 of the control device 109 may be described in detail herein.

[0049] In one implementation, one or more first modules 204 may include but are not limited to an intention recognition module 206 associated with the control device 109, an emission probability determination module 207, an extended wake-up duration estimation module 208, a mode configuration module 209 and one or more other modules 210.

[0050] In an embodiment, the first data 205 in the first memory 203 may include voice input data 211 (also referred to as at least one voice input 211), intention data 212 (also referred to as intention 212), probability data 213 (also referred to as probability of issuance 213), historical data 214, contextual factors 215 (also referred to as one or more contextual factors 215), extended wake-up duration data 216 (also referred to as extended wake-up duration 216), mode configuration data 217, predetermined thresholds 218 and other data 219 associated with the control device 109.

[0051] In an embodiment, the first data 205 in the first memory 203 may be processed by one or more first modules 204 of the control device 109. In an embodiment, one or more first modules 204 may be implemented as a dedicated unit, and when implemented in this manner, the module may be configured to have the functions defined in the present disclosure to produce novel hardware. As used herein, the term module may refer to an application-specific integrated circuit (ASIC), an electronic circuit, a field programmable gate array (FPGA), a programmable system on chip (PSoC), a combinational logic circuit, and / or other suitable components that provide the functions.

[0052] The control device 109 may be configured to control the operation of the voice assistant device 101 based on at least one voice input 211 received by the voice assistant device 101. The at least one voice input 211 may be processed by the control device 109 to control the voice assistant device 101. In an embodiment, each of the voice inputs provided to the voice assistant device 101 may be dynamically provided to the control device 109 to dynamically control the voice assistant device 101.

[0053] Upon receiving at least one voice input 211, the intent recognition module 206 of the control device 109 may be configured to recognize the intent 212 of the at least one voice input 211. In an embodiment, the intent 212 of the at least one voice input 211 may be recognized by performing voice recognition, voice synthesis, NLU on the at least one voice input 211. In an embodiment, the conversational knowledge of previous voice inputs of the voice assistant device 101 may be used to determine the intent 212 of the received at least one voice input 211. One or more other processing techniques known to those skilled in the art may be implemented in the intent recognition module 206 to recognize the intent 212 of the at least one voice input 211.

[0054] In addition, the emission probability receiving module 207 may be configured to receive an emission probability 213 of a subsequent speech input from the probability determination device 110. In an embodiment, the probability determination device 110 may be one of the other modules 210 of the control device 109. In an embodiment, the probability determination device 110 may be connected to the control device 109 via a communication network. The probability determination device 110 may determine the emission probability 213, and the emission probability receiving module 207 may be configured to communicate with the probability determination device 110 to receive the determined emission probability 213.

[0055] Figure 3 A detailed block diagram of the probability determination device 110 for determining the emission probability 213 according to some embodiments of the present disclosure is shown.

[0056] The probability determination device 110 may include a second processor 301, a second I / O interface 302, and a second memory 303. In some embodiments, the second memory 303 may be communicatively coupled to the second processor 301. The second memory 303 stores instructions executable by the second processor 301, which when executed may cause the probability determination device 110 to determine the probability of issuing 213. In an embodiment, the second memory 303 may include one or more second modules 304 and second data 305. The one or more second modules 304 may be configured to use the second data 305 to perform the steps of the present disclosure to determine the probability 213. In an embodiment, each of the one or more second modules 304 may be a hardware unit that may be external to the second memory 303 and coupled to the probability determination device 110. The probability determination device 110 may be implemented in various computing systems, such as a laptop, a desktop computer, a personal computer (PC), a notebook, a smart phone, a tablet computer, an e-book reader, a server, a network server, etc.

[0057] The second data 305 and the one or more second modules 304 in the second memory 303 of the probability determination device 110 may be described in detail herein.

[0058] In one implementation, the one or more second modules 304 may include but are not limited to a keyword extraction module 306 , a domain identification module 307 , a domain matching score calculation module 308 , a probability determination module 309 , and one or more other modules 310 associated with the probability determination device 110 .

[0059] In an embodiment, the second data 305 in the second memory 303 may include speech input data 311 (also referred to as at least one speech input 311), keyword data 312 (also referred to as one or more keywords 312), domain data 313 (also referred to as domain 313), domain matching score data 314 (also referred to as domain matching score 314), probability data 315 (also referred to as probability of issuance 315), and other data 316 associated with the probability determination device 110. In an embodiment, the speech input data 211 in the first memory 203 of the control device 109 may be the same as the speech input data 311 in the second memory 303 of the probability determination device 110. In an embodiment, the probability data 214 in the first memory 203 of the control device 109 may be the same as the probability data 315 in the second memory 303 of the probability determination device 110.

[0060] In an embodiment, the second data 305 in the second memory 303 may be processed by one or more second modules 304 of the probability determination device 110. In an embodiment, one or more second modules 304 may be implemented as a dedicated unit, and when implemented in this manner, the module may be configured to have the functions defined in the present disclosure to produce novel hardware. As used herein, the term module may refer to an application-specific integrated circuit (ASIC), an electronic circuit, a field programmable gate array (FPGA), a programmable system on chip (PSoC), a combinational logic circuit, and / or other suitable components that provide the functions.

[0061] The keyword extraction module 306 may be configured to extract one or more keywords 312 from a plurality of words in at least one speech input 311. In an embodiment, the keyword extraction module 306 may be configured to recognize a plurality of words in at least one speech input 311. Once the plurality of words are recognized, each of the plurality of words is assigned a weight. In an embodiment, a weight is assigned to each of the plurality of words based on the intent 212 of the at least one speech input 311. For example, a higher weight may be assigned to a word that is closely related to the intent 212 of the at least one speech input 311, and a lower weight may be assigned to a word that is least related to the intent 212. In addition, the keyword extraction module 306 may be configured to recognize a word having a weight greater than a predetermined weight value as one or more keywords 312 of the at least one speech input 311. Such one or more keywords 312 represent the at least one speech input 311. The domain identification module 307 may be configured to use the one or more keywords 312 to identify a domain 313 associated with the at least one speech input 311. In an embodiment, the probability determination device 110 may be associated with a library that includes a plurality of pre-stored domains associated with the speech assistance device 101. In an embodiment, a plurality of pre-stored domains may be identified and stored in a library based on a previous voice input or request from at least one user 102. In an embodiment, the plurality of pre-stored domains may include entertainment, news, sports, music, appliance configuration, etc. In addition, each of the plurality of pre-stored domains may be associated with a corresponding keyword. When one or more keywords 312 are extracted from at least one voice input 311, the one or more keywords 312 may be matched with keywords of the plurality of pre-stored domains. Through this matching, the domain 313 associated with the at least one voice input 311 may be identified by the domain identification module 307.

[0062] In addition, the domain matching score calculation module 308 may be configured to calculate a domain matching score 314 for the identified domain 313. In an embodiment, the domain matching score 314 may be calculated for the domain 313 using each of a plurality of pre-stored domains. In addition, based on the domain matching score 314, the probability determination module 309 of the probability determination device 110 may be configured to determine a probability 315 of issuing a subsequent voice input to the voice assistant device 101. In addition, in order to determine the probability 315, one or more contextual factors 215 and historical data 214 are also considered together with the domain matching score 314. In an embodiment, the historical data 214 may include past requests of at least one user 102 and corresponding responses provided by the voice assistant device 101. In an embodiment, the historical data 214 may include a time of request, a date of request, a location of request, and other data related to past requests from at least one user 102. In an embodiment, the one or more contextual factors 215 include at least one of a user-related factor, a time-related factor data, and an environment-related factor. For example, consider that at least one user 102 has a habit of triggering the voice assistant device 101 and giving commands to retrieve information related to weather, news, and traffic upon waking up. In this case, upon receiving the initial command from the at least one user 102 in the morning, the probability 315 of issuing subsequent voice inputs may be determined to be high. Similarly, consider that at least one user 102 continuously asks questions related to a single domain. Since there is a possibility that the at least one user 102 may ask other questions related to the single domain, the probability 315 may be determined to be high. In addition, consider a command that a user turns on a TV and then makes a phone call. In this case, the probability 315 may be determined to be low because the possibility of receiving further commands from the at least one user 102 is low.

[0063] In an embodiment, the other modules 311 of the probability determination device 110 may include a dialogue knowledge establishment module, which is configured to establish dialogue knowledge for at least one speech input 311 based on one or more extracted keywords 313. Such dialogue knowledge can be established for each dialogue of at least one user 102 using the voice assistance device 101. In an embodiment, the dialogue knowledge can be stored in a library.

[0064] In an embodiment, the intent recognition module 206 may be configured to use conversational knowledge of a previous conversation of at least one user 102 to recognize the intent 212 of the at least one voice input 211 .

[0065] The probability of emission 315 is determined by the probability determination device 110 and provided to the control device 109. Based on the probability 213, the extended wake-up duration estimation module 208 can be configured to determine an extended wake-up duration 2167 for the voice assistant device 101. First, the probability 213 is compared with the predetermined threshold 218. In the case where the determined probability 213 is greater than the predetermined threshold 218, an extended wake-up duration 216 of the voice assistant device 101 can be determined. In addition, the duration of the wake-up mode of the voice assistant device 101 can be extended to the extended wake-up duration 216 to receive a subsequent voice input from the at least one user 102. In an embodiment, the predetermined threshold 218 can be determined by a device connected to the voice assistant device 101 by analyzing the historical data 214. For example, if the at least one user 102 frequently has a deep conversation with the voice assistant device 101, the predetermined threshold 218 can be selected as a lower value. Even if the value of the probability 213 is small, the extended wake-up duration 216 can be determined. Similarly, consider that the at least one user 102 mainly uses the voice assistant device 101 to control home appliances, and less for retrieving news updates. When the intent 212 is related to news updates, the predetermined threshold 218 may be set to a higher value because a higher probability is required to extend the wake-up duration.

[0066] In an embodiment, the extended wake-up duration 216 may be estimated to be proportional to the probability 213 of the subsequent speech input being emitted. For example, if the determined probability 213 is high, the extended wake-up duration 216 may be determined to last for a longer duration. Similarly, when the determined probability 213 is low but equal to or greater than the predetermined threshold 218, the extended wake-up duration 216 may be determined to last for a shorter duration.

[0067] In addition, upon determining the extended wake-up duration 216, the mode configuration module 209 may configure the voice assistant device 101 to operate in the wake-up mode during the extended wake-up duration 216. In an embodiment, when the probability 213 is determined to be less than the predetermined threshold 218, the mode configuration module 209 may configure the voice assistant device 101 to operate in the sleep mode until a trigger to operate in the wake-up mode is detected. In addition, during the extended wake-up duration 216, if no subsequent voice input is detected, the mode configuration module 209 may configure the voice assistant device 101 to operate in the sleep mode. The mode in which the voice assistant device 101 is to be operated is stored as the mode configuration data 217. The mode configuration data 217 may be one of the sleep mode and the wake-up mode.

[0068] Other data 220 and 317 may store data generated by modules for performing various functions of the control device 109 and the probability determination device 110, including temporary data and temporary files. The one or more first modules 204 and the one or more second modules 304 may also include other modules 211 and other modules 311, respectively, to perform various different functions of the control device 109 and the probability determination device 110. It should be understood that such modules may be represented as a single module or a combination of different modules.

[0069] Figure 4a and Figure 4b An exemplary schematic diagram for controlling a voice assistant device 101 according to some embodiments of the present disclosure is shown. Consider that a user 401 is associated with a voice assistant device 402. The voice assistant device 402 may be controlled by a control device 403 using a probability determination device 404.

[0070] First, the voice assistant device 402 may operate in a sleep mode. When a trigger and a first command are received from the user 401, the voice assistant device 402 may operate in a wake-up mode. The control device 403 receives the first command, which is provided for determining the probability of issuance by the probability determination device 404. When the probability is greater than a predetermined threshold, an extended wake-up duration T relative to the first command is determined. 1 In response to the first command, the wake-up mode of the voice assistant device 402 is extended to the extended wake-up duration T 1 Some use cases associated with controlling the voice-assisted device 402 are described below.

[0071] consider Figure 5a. Consider that user 501 has a habit of requesting news updates from voice assistant device 502 in the morning. User 501 activates the service of voice assistant device 502 by providing a voice input such as "(trigger word), good morning". The voice input is received by voice assistant device 502 and provided to control unit 503 including probability determination means (not shown in the exemplary embodiment). Since the historical data includes a request for further requests from user 502, the probability of issuance of subsequent voice inputs can be determined to be a higher value. Therefore, it can be determined that the wake-up duration is extended for a longer duration. For example, in this case, the extended wake-up duration can be 60 seconds. Using natural language generation (NLG), a response to the voice input can be provided as "Good morning, have a nice day". While responding, the wake-up duration can also be extended to 60 seconds. As predicted, user 501 can provide a subsequent voice input such as "read news updates". The voice input is received by voice assistant device 502 and provided to control means 503. Since historical data of user 501 indicates that user 501 typically only requests news updates in the morning, and news updates have been provided to user 501, the probability may be low. Therefore, the extended wake-up duration may last for a shorter duration. For example, the extended wake-up duration involving subsequent voice input may be 20 seconds. Using NLG, a response to the voice input may be provided as "Here is the news update." While responding, the wake-up duration may also be extended to 20 seconds. If no voice input is received within 20 seconds, the voice assistant device 502 may be configured to be in sleep mode.

[0072] consider Figure 5b. Consider that a user has a habit of controlling home appliances as soon as he enters his home from the office in the evening. User 501 starts the service of voice assistant device 502 by providing voice input "(trigger word), turn on the light". By receiving the voice input through voice assistant device 502 and providing the voice input to control device 503, and the historical data includes requesting further requests, the probability of issuing subsequent voice inputs can be determined to be a higher value, and the extended wake-up duration can also be determined to last for a longer duration. For example, in this case, the extended wake-up duration can be 60 seconds. Voice assistant device 502 can turn on the light and can use NLG to provide a response to the voice input as "Welcome home, I will turn on the light". While responding, the wake-up duration can be extended to 60 seconds. As predicted, user 501 provides subsequent voice input such as "Turn on the AC and set the temperature to 20". Voice input is received through voice assistant device 502 and provided to control device 503. Since the historical data of user 501 indicates that multiple home appliances are controlled at night, the probability may be high. Thus, the extended wake-up duration may last for a longer duration, but be shorter than the previous wake-up duration. For example, the extended wake-up duration associated with the subsequent voice input may be 40 seconds. The voice assistant device 502 may turn on the AC and set the temperature to 20. Furthermore, a response to the voice input may be provided using NLG as "the AC is on and the temperature is set to 20". While responding, the wake-up time may also be extended to 40 seconds, and further voice input from the user 501 may be awaited within the span of 40 seconds. If no voice input is received within the 40 seconds, the voice assistant device 502 may be configured to be in sleep mode.

[0073] consider Figure 5c, wherein a user 501 conducts a deep conversation with a voice assistant device 502. The deep conversation may include multiple questions from the user 501 related to a single context. The user 501 starts the service of the voice assistant device 502 by providing a voice input "(trigger word), who is Donald Trump". The voice input is received by the voice assistant device 502 and provided to the control device 503. The voice input may be a new context about a request to the user in the past, and therefore, the determined probability may be a lower value. However, if the user 501 has previously conducted a deep conversation with the voice assistant device 502, the probability may be greater than a predetermined threshold. Therefore, the extended wake-up duration may last for a shorter period of time. For example, in this case, the extended wake-up duration may be 20 seconds. The voice assistant device 502 may use NLG and provide a response to the voice input as "Donald Trump is the 45th and current president of the United States". While responding, the wake-up time may also be extended to 20 seconds. The user 501 provides a subsequent voice input as "How old is he?" within 20 seconds. The voice input is received by the voice assistant device 502 and provided to the control device 503. Since the subsequent voice input is related to the previous voice input, the probability of receiving the subsequent voice input can be determined as a high value. Therefore, the extended wake-up duration can be longer than the previous wake-up mode duration. For example, the extended wake-up duration associated with the subsequent voice input can be 40 seconds. The response to the voice input can be provided as "He was born in 1946 and he is 72 years old" using NLG. While responding, the wake-up time can also be extended to 40 seconds, and further voice input from the user 501 is waited for within the span of 40 seconds. If no voice input is received within the 40 seconds, the voice assistant device 502 can be configured to be in sleep mode. Within the 40-second duration, the user 501 provides the subsequent voice input as "Who was the president before him." The voice input is received by the voice assistant device 502 and provided to the control device 503. In addition, since the subsequent voice input is related to the previous voice input, the probability of receiving the subsequent voice input can be determined as a high value. Therefore, the extended wake-up duration can be longer than the previous wake-up mode duration. For example, the extended wake-up duration associated with subsequent voice input may be 60 seconds. The response to the voice input may be provided using NLG as "Barack Obama was the president before Donald Trump". While responding, the wake-up time may also be extended to 60 seconds, and further voice input from the user 501 may be awaited within the 60-second span. If no voice input is received within the 60 seconds, the voice assistant device 502 may be configured to be in a sleep mode.

[0074] Refer to Figure 4a and Figure 4b, consider that user 401 is immersed in a continuous conversation and is in the extended wake-up duration T 1 In this case, the continued dialog is received by the voice assistant device 402 together with the second command and provided to the control device 403 and the probability determination device 404. When the probability determined for the second command is greater than a predetermined threshold, the extended wake-up duration T relative to the second command is determined. 2 In response to the second command, the wake-up mode of the voice assistant device 402 is extended to the extended wake-up duration T 2 Consider further that the user 401 is only immersed in the ongoing conversation without issuing any command to the voice assistant device 502. 2 Based on the intention of the continued conversation, a probability of the continued conversation can be determined. If the continued conversation is related to the context of the previous command, the probability can be higher than a predetermined threshold. If the continued conversation is not related to the context of the previous command, the probability can be lower than the predetermined threshold. Considering that the probability is higher than the predetermined threshold, an extended wake-up duration T for the second command is determined. 3 The wake-up mode of the voice assistant device 502 is extended to the extended wake-up duration T 3 .

[0075] consider Figure 5d, wherein a user 501 initiates a service of a voice assistant device 502 by providing a voice input "(trigger word), any notifications to me". The voice input is received by the voice assistant device 502 and provided to the control device 503. The voice input may be associated with a one-time request, and therefore, the determined probability may be a lower value. Therefore, the extended wake-up duration may last for a shorter period of time. For example, in this case, the extended wake-up duration may be 20 seconds. The voice assistant device 502 may use NLG and provide a response to the voice input as "You received a message from Anna saying that we will have dinner together tonight". While responding, the wake-up time may also be extended to 20 seconds. The user 501 provides a subsequent voice input as a continuous conversation, such as "Send her a message saying, sorry, I have an office dinner meeting planned tonight, I can't attend tonight, how about tomorrow?". The voice input is a continuous conversation that lasts more than 20 seconds. In addition, since the subsequent voice input is related to the previous voice input, the probability of receiving the subsequent voice input may be determined to be a high value. Therefore, the extended wake-up duration may last for a longer duration than the previous wake-up mode duration. For example, the extended wake-up duration associated with the subsequent voice input may be 60 seconds. However, if voice input is still received from the user 501, the control device 503 may be configured to further extend the wake-up mode duration of the voice assistance device 502. The process of extending the wake-up mode duration may continue until the voice input is received and there is no voice input within the predetermined duration. For example, first, the wake-up duration is extended to 60 seconds and further extended to 20 seconds. Upon completion of the extended wake-up duration, a response to the voice input may be provided as "OK, I send the message" using NLG. The voice assistance device 502 may be configured to be in sleep mode when sending the response.

[0076] consider Figure 5e, in which a first user 501.1 and a second user 501.2 are having a conversation. The first user 501.1 may talk to the second user 501.2 and say "Do you want to go to a movie?" The second user 501.2 may reply "Sure, which movie do you want to watch?" Since neither the first user 501.1 nor the second user 501.2 said a trigger word, the voice assistant device 502 may still be in sleep mode. In addition, the first user 501.1 starts the service of the voice assistant device 502 by saying "(trigger word), which English movies are playing in the theater now?". The voice input is received by the voice assistant device 502 and provided to the control device 503. The probability of the voice input may be determined to be greater than a predetermined threshold, and the extended wake-up duration may be determined to be 30 seconds. The voice assistant device 502 may use NLG and provide a response to the voice input as "Some English movies playing in the theater now are: Avengers: Endgame, Captain Marvel, Detective Pikachu, Aladdin, Godzilla". While responding, the wake-up time may also be extended to 30 seconds. Upon hearing the response, the first user 501.1 engages in a conversation with the second user 501.2 saying "Which movie do you want to watch?" The second user 501.2 may reply to the first user 501.1 saying "Okay, let's go see Avengers: Endgame." During the conversation between the first user 501.1 and the second user 501.2, the voice assistant device 502 may be configured to receive voice inputs from the first user 501.1 and the second user 501.2, and provide the voice inputs to the control device 503. During the conversation, since each of the voice inputs is associated with an initial request from the first user 501.1, the control device 503 may determine that the probability is greater than a predetermined threshold. In addition, the wake-up duration of the voice assistant device 502 may be increased until a further request of the voice assistant device 502 is received. For example, as shown in the figure, the wake-up mode may be first extended to 20 seconds, and then further extended to 20 seconds and 40 seconds. When the first user 501.1 says "Book your movie tickets for the 6:30pm showing of Avengers: Endgame", the voice assistant device 502 may detect the voice input as a request and respond accordingly. The voice assistant device 502 may use NLG to book movie tickets and may respond by saying "Your movie tickets for the 6:30pm showing of Avengers: Endgame have been booked". Since the task is completed, the voice assistant device 502 may be configured to be in a sleep mode while sending the response.

[0077] Refer again Figure 4a and Figure 4b , consider that user 401 is in a continuous conversation and extends the wake-up duration T 1In this case, the continued dialog together with the second command is received by the voice assistant device 402 and provided to the control device 403 and the probability determination device 404. When the probability determined for the second command is less than a predetermined threshold, the extended wake-up duration may be determined to be 0. In response to the second command, the voice assistant device 402 may be configured to operate in a sleep mode.

[0078] consider Figure 5f . Consider that user 501 has a habit of playing music in the vehicle using voice assistant device 502 as soon as he enters the vehicle. User 501 starts the service of voice assistant device 502 by providing voice input "(trigger word), navigate to the airport". The voice input is received by voice assistant device 502 and provided to control device 503. Since the request for music has not been received, the probability of the issuance of the subsequent voice input can be determined to be a higher value, and the extended wake-up duration can be determined to last for a longer duration. For example, in this case, the extended wake-up duration can be 60 seconds. Voice assistant device 502 can provide navigation to the airport, and a response to the voice input can be provided as "OK, I start navigating to Bangalore International Airport" using NLG. While responding, the wake-up time can also be extended to 60 seconds. As predicted, user 501 provides "play some music" as a subsequent voice input. The voice input is received by voice assistant device 502 and provided to control device 503. Since the expected request has been received, the probability can be determined to be less than a predetermined threshold. Therefore, the extended wake-up duration can be determined to be 0. The response may be provided using NLG as “OK, play the latest Bollywood music.” Additionally, the voice assistant device 502 may be configured to be in a sleep mode.

[0079] Figure 6a A flow chart illustrating an exemplary method 600 of a voice assistant device 101 is shown in accordance with some implementations of the present disclosure.

[0080] At block 601, the voice assistant device 101 may be configured to receive at least one voice input from at least one user 102 during a wake-up mode of the voice assistant device. In an embodiment, the at least one voice input may be received dynamically from the voice assistant device 101.

[0081] At block 602, the voice assistant device 101 may be configured to recognize an intent associated with at least one voice input from at least one user 102. In an embodiment, the intent may be recognized using conversational knowledge of previous voice inputs from at least one user 102.

[0082] At block 603, the voice assistant device 101 determines the probability of the subsequent voice input being emitted. The probability of the subsequent voice input being emitted may be determined based on the intent, historical data, and one or more contextual factors. Figure 6b A flow chart of an exemplary method 603 for determining an utterance probability of a subsequent speech input according to some embodiments of the present disclosure is shown.

[0083] At block 606, the voice assistant device 101 may be configured to extract one or more keywords from a plurality of words in the at least one voice input based on the intent of the voice input. To extract the one or more keywords, a weight is assigned to each of the plurality of words in the at least one voice input. In addition, keywords in the plurality of words having a weight greater than a predetermined weight value are identified as the one or more keywords.

[0084] At block 607, the voice assistant device 101 may be configured to identify a domain associated with the at least one voice input based on the one or more keywords. One or more techniques known to those skilled in the art may be implemented to identify the domain.

[0085] At block 608, the voice assistant device 101 may be configured to calculate a domain matching score for the identified domain using a plurality of pre-stored domains associated with the at least one user 102. In an embodiment, the plurality of pre-stored domains may be identified based on previous voice input from the at least one user 102.

[0086] At block 609 , the voice assistant device 101 may be configured to determine a probability of utterance of a subsequent voice input to the voice assistant device 101 based on the domain matching score associated with the at least one user 102 , one or more contextual factors, and historical data.

[0087] Back to reference Figure 6a At block 604, when the probability is greater than the predetermined threshold, the voice assistant device 101 may be configured to estimate the extended wake-up duration of the voice assistant device. In an embodiment, when the probability is less than the predetermined threshold, the extended wake-up duration may be determined to be 0.

[0088] At block 605, the voice assistant device 101 may be configured to extend the duration of the wake-up mode of the voice assistant device 101 to the extended wake-up duration to receive subsequent voice input from the at least one user 102. When it is determined that the probability is less than a predetermined threshold, the voice assistant device 101 may be configured to be in a sleep mode until a trigger to operate in the wake-up mode is detected.

[0089] Figure 6a and Figure 6bThe method shown in may include one or more blocks for performing a process in at least one of the speech assistant device 101, the control device 109, and the probability determination device 110. The method may be described in the general context of computer executable instructions. Figure 6a and Figure 6b Generally, computer-executable instructions may include routines, programs, objects, components, data structures, procedures, modules, and functions that perform specific functions or implement specific abstract data types.

[0090] describe Figure 6a and Figure 6b The order of the method shown in the description is not intended to be construed as limiting, and any number of the described method blocks may be combined in any order to implement the method. In addition, individual blocks may be deleted from the method without departing from the scope of the subject matter described herein. Furthermore, the method may be implemented in any suitable hardware, software, firmware, or a combination thereof.

[0091] Computing System

[0092] Figure 7 A block diagram of an exemplary computer system 700 for implementing an embodiment consistent with the present disclosure is shown. In an embodiment, the computer system 700 is used to implement at least one of the speech assistance device 101, the control device 109, and the probability determination device 110. The computer system 700 may include a central processing unit (CPU or processor) 702. The processor 702 may include at least one data processor for executing processes in a virtual storage area network. The processor 702 may include a special processing unit, such as an integrated system (bus) controller, a memory management control unit, a floating point unit, a graphics processing unit, a digital signal processing unit, etc.

[0093] The processor 702 may be configured to communicate with one or more input / output (I / O) devices 709 and an input / output (I / O) device 710 via the I / O interface 701. The I / O interface 701 may employ communication protocols / methods such as, but not limited to, audio, analog, digital, mono, RCA, stereo, IEEE-1394, serial bus, Universal Serial Bus (USB), infrared, PS / 2, BNC, coaxial, component, composite, digital visual interface (DVI), high-definition multimedia interface (HDMI), RF antenna, S-video, VGA, IEEE802.n / b / g / n / x, Bluetooth, cellular such as code division multiple access (CDMA), high-speed packet access (HSPA+), global system for mobile communications (GSM), long-term evolution (LTE), WiMax, etc.

[0094] Using the I / O interface 701, the computer system 700 can communicate with one or more I / O devices 709 and I / O devices 710. For example, the input device 709 can be an antenna, a keyboard, a mouse, a joystick, an (infrared) remote control, a camera, a card reader, a fax machine, a dongle, a biometric reader, a microphone, a touch screen, a touch pad, a trackball, a stylus, a scanner, a storage device, a transceiver, a video device / source, etc. The output device 710 can be a printer, a fax machine, a video display such as a cathode ray tube (CRT), a liquid crystal display (LCD), a light emitting diode (LED), a plasma, a plasma display panel (PDP), an organic light emitting diode display (OLED), etc., an audio speaker, etc.

[0095] In some embodiments, the computer system 700 may be composed of at least one of the speech assistant device 101, the control device 109, and the probability determination device 110. The processor 702 may be configured to communicate with the communication network 711 via the network interface 703. The network interface 703 may communicate with the communication network 711. The network interface 703 may use a connection protocol, including but not limited to direct connection, Ethernet (e.g., twisted pair 10 / 100 / 1000BaseT), transmission control protocol / Internet protocol (TCP / IP), token ring, IEEE 802.11a / b / g / n / x, etc. The communication network 711 may include but is not limited to direct interconnection, local area network (LAN), wide area network (WAN), wireless network (e.g., using wireless application protocol), Internet, etc. Using the network interface 703 and the communication network 711, the computer system 700 may communicate with the user 712 to provide services to the user 712. The network interface 703 may employ connection protocols including, but not limited to, direct connection, Ethernet (eg, twisted pair 10 / 100 / 1000BaseT), Transmission Control Protocol / Internet Protocol (TCP / IP), Token Ring, IEEE 802.11a / b / g / n / x, etc.

[0096] The communication network 711 includes, but is not limited to, direct interconnection, e-commerce networks, peer-to-peer (P2P) networks, local area networks (LANs), wide area networks (WANs), wireless networks (e.g., using wireless application protocols), the Internet, Wi-Fi, etc. The first network and the second network may be dedicated networks or shared networks, which represent associations of different types of networks that communicate with each other using various protocols such as Hypertext Transfer Protocol (HTTP), Transmission Control Protocol / Internet Protocol (TCP / IP), Wireless Application Protocol (WAP), etc. In addition, the first network and the second network may include various network devices, including routers, bridges, servers, computing devices, storage devices, etc.

[0097] In some embodiments, the processor 702 may be configured to communicate with the memory 705 (eg, Figure 7 The storage interface 704 may communicate with the memory 705 using connection protocols such as Serial Advanced Technology Attachment (SATA), Integrated Drive Electronics (IDE), IEEE-1394, Universal Serial Bus (USB), Fibre Channel, Small Computer System Interface (SCSI), etc. The memory 705 includes but is not limited to a memory drive, a removable disk drive, etc. The memory drive may also include a drum, a disk drive, a magneto-optical drive, an optical drive, a redundant array of independent disks (RAID), a solid-state memory device, a solid-state drive, etc.

[0098] The memory 705 may store a collection of program or database components, including, but not limited to, a user interface 706, an operating system 707, a web browser 708, etc. In some embodiments, as described in the present disclosure, the computer system 700 may store user / application data 706, such as data, variables, records, etc. Such a database may be implemented as a fault-tolerant, relational, extensible, secure database, e.g. or

[0099] Operating system 707 can facilitate resource management and operation of computer system 700. Examples of operating systems include, but are not limited to, APPLE OS X, UNIX-like distributions (e.g., BERKELEY SOFTWARE DISTRIBUTION TM (BSD), FREEBSD TM , NETBSD TM 、OPENBSD TM etc.), LINUX DISTRIBUTIONS TM (For example, RED HAT TM 、UBUNTU TM 、KUBUNTU TM etc.), IBM TM OS / 2, MICROSOFT TM WINDOWS TM (XP TM VISTA TM / 7 / 8, 10, etc.), iOS TM , ANDROID TM , OS, etc.

[0100] In some embodiments, the computer system 700 may implement a program component stored by a web browser 708. The web browser 708 may be a hypertext viewing application such as Microsoft Internet Explorer, Google Chrome, Mozilla Firefox, Apple Safari, etc. Secure web browsing may be provided using secure hypertext transfer protocol (HTTPS), secure link layer (SSL), transport layer security (TLS), etc. The web browser 508 may utilize tools such as AJAX, DHTML, Adobe Flash, JavaScript, Java, application programming interface (API), etc. In some embodiments, the computer system 700 may implement a program component stored by a mail server. The mail server may be an Internet mail server such as Microsoft Exchange, etc. The mail server may utilize tools such as ASP, ActiveX, ANSI C++ / C#, Microsoft.NET, CGIscripts, Java, JavaScript, PERL, PHP, Python, WebObjects, etc. The mail server may utilize communication protocols such as Internet Message Access Protocol (IMAP), Messaging Application Programming Interface (MAPI), Microsoft Exchange, Post Office Protocol (POP), Simple Mail Transfer Protocol (SMTP), etc. In some implementations, the computer system 700 may implement a program component stored in a mail client. The mail client may be an email viewing application such as Apple Mail, Microsoft Entourage, Microsoft Outlook, Mozilla Thunderbird, etc.

[0101] In addition, one or more computer-readable storage media can be used to implement embodiments consistent with the present disclosure. Computer-readable storage media refers to any type of physical memory on which processor-readable information or data can be stored. Therefore, a computer-readable storage medium can store instructions executed by one or more processors, including instructions for causing the processor to perform steps or stages consistent with the embodiments described herein. The term "computer-readable medium" should be understood to include tangible items and exclude carrier waves and transient signals, i.e., non-transient. Examples include random access memory (RAM), read-only memory (ROM), volatile memory, non-volatile memory, hard disk drive, CD ROM, DVD, flash drive, disk, and any other known physical storage medium.

[0102] The described operations may be implemented as methods, systems or products that use standard programming and / or engineering techniques to produce software, firmware, hardware or any combination thereof. The described operations may be implemented as codes maintained in a "non-transitory computer-readable medium", where a processor may read and execute the code from a computer-readable medium. The processor is at least one of a microprocessor and a processor that can process and execute inquiries. Non-transitory computer-readable media may include media such as magnetic storage media (e.g., hard disk drives, floppy disks, tapes, etc.), optical storage (CD-ROMs, DVDs, optical disks, etc.), volatile storage devices, and non-volatile storage devices (e.g., EEPROMs, ROMs, PROMs, RAMs, DRAMs, SRAMs, flash memory, firmware, programmable logic, etc.). In addition, non-transitory computer-readable media may include all computer-readable media except temporary ones. The code that implements the described operations may also be implemented in hardware logic (e.g., integrated circuit chips, programmable gate arrays (PGAs), application-specific integrated circuits (ASICs), etc.).

[0103] An "industrial product" includes a non-transitory computer-readable medium and / or hardware logic in which code may be implemented. A device in which code implementing an embodiment of the described operations is encoded may include a computer-readable medium or hardware logic. Of course, those skilled in the art will recognize that many modifications may be made to this configuration without departing from the scope of the present invention, and that an industrial product may include suitable information bearing media known in the art.

[0104] The terms "an embodiment," "embodiment," "embodiments," "the embodiment," "the embodiments," "one or more embodiments," "some embodiments," and "one embodiment" mean "one or more (but not all) embodiments of the invention," unless expressly stated otherwise.

[0105] The terms "including," "comprising," "having" and variations thereof mean "including but not limited to," unless expressly stated otherwise.

[0106] Unless expressly stated otherwise, the listing of enumerated items does not imply that any item or all of the categories of items are mutually exclusive.

[0107] The terms "a", "an" and "the" mean "one or more" unless expressly stated otherwise.

[0108] The description of an embodiment with multiple components in communication with each other does not imply that all such components are required. Rather, various optional components are described to illustrate the many possible embodiments of the present invention.

[0109] When describing a single device or product in this article, it is obvious that more than one device / product (whether they cooperate or not) can be used to replace a single device / product. Similarly, in the case of describing more than one device or product (whether they cooperate or not) in this article, it is obvious that a single device / product can be used to replace more than one device or product, or different numbers of devices / products can be used to replace the devices or programs of the number shown. The function and / or feature of the device can be optionally implemented by one or more other devices that are not clearly described as having such function / feature. Therefore, other embodiments of the present invention do not need to include the device itself.

[0110] Figure 6a and Figure 6b The operations shown in the figure illustrate specific events that occur in a specific order. In an optional embodiment, some operations can be performed, modified or removed in different orders. In addition, steps can be added to the above logic and still meet the described embodiments. In addition, the operations described herein can occur sequentially, or some operations can be processed in parallel. In addition, the operations can be performed by a single processing unit or by a distributed processing unit.

[0111] Finally, the language used in this specification is selected primarily for readability and instructional purposes, and may not be selected to delimit or limit the subject matter of the present invention. Therefore, it is intended that the scope of the present invention is not limited by this specific embodiment, but rather by any claims issued based on the application hereof. Therefore, the disclosure of the embodiments of the present invention is intended to illustrate rather than limit the scope of the present invention, which is set forth in the appended claims.

[0112] Although various aspects and embodiments have been disclosed herein, other aspects and embodiments will be apparent to those skilled in the art. The various aspects and embodiments disclosed herein are for purposes of illustration and not for purposes of limitation, with the true scope and spirit being indicated by the appended claims.

Claims

1. A method performed by a voice assistant device, the method comprising: receiving at least one voice input from at least one user via a voice assistant device during a wake-up mode of the voice assistant device; identifying, by the voice assistant device, an intent associated with the at least one voice input from the at least one user; Determining, by the speech assistant device, a probability of an utterance of a subsequent speech input from the at least one user based on at least one of the following: The intention, Historical data, and one or more contextual factors; comparing the probability and a threshold value by the speech assistance device; as well as When the probability is greater than the threshold, the duration of the wake-up mode of the voice assistant device is extended by the voice assistant device to extend the wake-up duration to receive the subsequent voice input from the at least one user.

2. The method according to claim 1, further comprising: When the probability is greater than the threshold, an extended wake-up duration of the voice assistant device is determined by the voice assistant device.

3. The method according to claim 1, wherein: The threshold is determined by analyzing the historical data via a device connected to the voice assistance device.

4. The method according to claim 1, further comprising: Determining, by a control device, that the probability is less than a threshold value; as well as The voice assistant device is configured by the control device to operate in the sleep mode until a trigger to operate in the wake-up mode is detected.

5. The method according to claim 1, further comprising: If the subsequent voice input is not detected during the extended wake-up duration, the voice assistant device is configured by the control means to operate in a sleep mode.

6. The method according to claim 1, wherein: The intent associated with the at least one voice input is identified by performing natural language understanding on the at least one voice input.

7. The method according to claim 1, wherein: Determining the probability of the subsequent speech input being issued comprises: extracting one or more keywords from a plurality of words in the at least one speech input; identifying a domain associated with the at least one speech input based on the one or more keywords; calculating a domain matching score for the identified domain using a plurality of pre-stored domains associated with the user; and Based on the domain matching score, a probability of utterance of the subsequent speech input to the speech assistant device is determined.

8. The method according to claim 7, wherein: Extracting the one or more keywords includes: assigning a weight to each of the plurality of characters of the at least one speech input; and One or more keywords having weights greater than a predetermined weight value among the plurality of words are identified as the one or more keywords.

9. Voice assistance equipment, including: a receiver configured to receive at least one voice input from at least one user when operating in the wake-up mode; processor; as well as a memory communicatively coupled to the processor, The memory stores instructions that can be executed by the processor, and when executed by the processor, the instructions cause the processor to: identifying an intent associated with the at least one voice input from the user; Determining a probability of utterance of a subsequent speech input from the at least one user based on at least one of: The intention, Historical data, and one or more contextual factors; comparing the probability to a threshold; and When the probability is greater than the threshold, the duration of the wake-up mode is extended with an extended wake-up duration to receive the subsequent voice input from the at least one user.

10. The voice assistance device according to claim 9, wherein: The memory stores processor-executable instructions that enable the processor to perform the following steps: The extended wake-up duration is determined when the probability is greater than the threshold.

11. The voice assistance device according to claim 9, wherein: The predetermined threshold is determined by analyzing the historical data via a device connected to the voice assistance device.

12. The voice assistance device according to claim 9, wherein: The memory stores processor-executable instructions that enable the processor to perform the following steps: determining that the probability is less than a predetermined threshold; and The system is configured to operate in the sleep mode until a trigger to operate in the wake-up mode is detected.

13. The voice assistance device according to claim 9, wherein: The intent associated with the at least one speech input is determined by performing natural language understanding (NLU) on the at least one speech input.

14. The speech assistant device of claim 9, the memory storing processor-executable instructions that enable the processor to determine the probability of utterance of the subsequent speech input by performing the following steps: extracting one or more keywords from a plurality of words in the at least one speech input; identifying a domain associated with the at least one speech input based on the one or more keywords; calculating a domain matching score for the identified domain using a plurality of pre-stored domains associated with the user; as well as Based on the domain matching score, a probability of utterance of the subsequent speech input to the speech assistant device is determined.

15. The voice assistance device according to claim 14, wherein: The extraction of the one or more keywords is performed by the following steps: assigning a weight to each of the plurality of characters of the at least one speech input; as well as One or more keywords having weights greater than a predetermined weight value among the plurality of words are identified as the one or more keywords.

16. The voice assistance device according to claim 9, wherein: The memory stores processor-executable instructions that enable the processor to perform the following steps: Conversation knowledge is established for the at least one speech input based on the extracted one or more keywords, wherein the conversation knowledge is used to identify the intention of the subsequent speech input of the speech assistance device.

17. The voice assistant device according to claim 9, wherein: The memory stores processor-executable instructions that enable the processor to perform the following steps: During the extended wake-up duration, if the subsequent voice input is not detected, the voice assistant device is configured in a sleep mode.

Images