Voice broadcasting method and device, equipment and storage medium

Abstract

The invention discloses a voice broadcast method and device, equipment and a storage medium, relates to the technical field of artificial intelligence, and can solve the technical problem that a user cannot use a voice interaction function when electrical equipment is disconnected in the prior art, so that the use experience of the user can be improved. The method can be applied to electrical equipment, and comprises the steps of determining a target keyword sequence corresponding to a voice instruction under the condition that the voice instruction is detected; the target keyword sequence comprises at least one target key segmented word; searching a target audio corresponding to the target keyword sequence in a locally stored offline corpus, and performing voice broadcast on the searched target audio; wherein the offline corpus comprises candidate keyword sequences and candidate audios corresponding to the candidate keyword sequences respectively, and each candidate keyword sequence comprises at least one candidate key segmented word.

Description

Technical Field

[0001] The present application relates to the field of artificial intelligence technology, and in particular to a voice broadcasting method, device, equipment and storage medium. Background Art

[0002] With the rapid development of artificial intelligence technology, the application of human-computer voice interaction is becoming more and more widespread. At present, human-computer voice interaction has been widely used in electrical equipment. Taking washing equipment as an example, when using washing equipment, users can interact with the washing equipment through human-computer voice and obtain some professional knowledge related to washing through the voice broadcast of the washing equipment.

[0003] However, in the prior art, when a user performs human-computer voice interaction with an electrical device, the electrical device must be connected to the Internet. When the electrical device is disconnected from the Internet, the user will not be able to use the voice interaction function, which affects the user's experience. Summary of the invention

[0004] The present application provides a voice broadcast method, device, equipment and storage medium, which can solve the technical problem in the prior art that users cannot use the voice interaction function when the electrical device is disconnected from the network, thereby improving the user experience.

[0005] In order to achieve the above objectives, this application adopts the following technical solutions:

[0006] In a first aspect, the present application provides a voice broadcast method, which can be applied to electrical equipment, comprising: when a voice command is detected, determining a target keyword sequence corresponding to the voice command; the target keyword sequence includes at least one target key word; searching for a target audio corresponding to the target keyword sequence in a locally stored offline corpus, and voice broadcasting the found target audio; wherein the offline corpus includes candidate keyword sequences and candidate audio corresponding to each candidate keyword sequence, and each candidate keyword sequence includes at least one candidate key word.

[0007] In the technical solution provided by the present application, an offline corpus including each candidate keyword sequence and the candidate audio corresponding to each candidate keyword sequence can be pre-stored locally. When a voice command is detected, the target keyword sequence corresponding to the voice command can be determined first; then, the target audio corresponding to the target keyword sequence can be searched in the offline corpus, and the found target audio can be voice broadcast. It can be seen that the present application can realize human-computer voice interaction through the offline corpus pre-stored locally. Since the offline corpus is pre-stored locally and does not need to be obtained through the Internet, in the technical solution provided by the present application, the human-computer voice interaction between the user and the electrical device is not limited by the network status of the electrical device. When the electrical device is disconnected from the network, the user can still use the voice interaction function. Therefore, the present application can improve the user experience.

[0008] Optionally, in a possible design, after the above “searching for a target audio corresponding to the target keyword sequence in a locally stored offline corpus”, the voice broadcast method provided by the present application may include:

[0009] If the target audio corresponding to the target keyword sequence is not found in the offline corpus, determining whether the electrical device is in an online state;

[0010] When it is determined that the electrical device is in an online state, a target keyword sequence or a voice command is sent to the background server to instruct the background server to determine the target audio according to the target keyword sequence or the voice command;

[0011] The target audio is obtained from the backend server, and the target audio is voiced, and the correspondence between the target keyword sequence and the target audio is updated to the offline corpus.

[0012] Optionally, in another possible design, the above-mentioned “sending a target keyword sequence or voice instruction to the background server to instruct the background server to determine the target audio according to the target keyword sequence or voice instruction” may include:

[0013] The target keyword sequence is sent to the backend server to instruct the backend server to search for the target audio corresponding to the target keyword sequence from the online corpus.

[0014] Optionally, in another possible design, the above-mentioned “sending a target keyword sequence or voice instruction to the background server to instruct the background server to determine the target audio according to the target keyword sequence or voice instruction” may include:

[0015] Send a voice command to the background server to instruct the background server to process the voice command based on the preset voice recognition model and obtain the target audio output by the preset voice recognition model.

[0016] Optionally, in another possible design, searching for a target audio corresponding to the target keyword sequence in a locally stored offline corpus includes:

[0017] Based on the target keyword sequence, determine the inquiry type to which the voice command belongs;

[0018] Based on the query type, a target query sub-corpus is determined from each query sub-corpus of the offline corpus;

[0019] Search the target audio corresponding to the target keyword sequence in the target query subcorpus.

[0020] Optionally, in another possible design, the inquiry type includes at least an activation inquiry class and a fault inquiry class.

[0021] Optionally, in another possible design, searching for a target audio corresponding to the target keyword sequence in a locally stored offline corpus includes:

[0022] Analyze the voice command to determine the timbre characteristics of the voice command;

[0023] Determine the target user of the voice command based on the timbre characteristics;

[0024] Determine a target account sub-corpus corresponding to the target user from each account sub-corpus of the offline corpus;

[0025] Search the target audio corresponding to the target keyword sequence in the target account sub-corpus.

[0026] In a second aspect, the present application provides a voice broadcast device, which can be configured in an electrical device, including a determination module and a voice module;

[0027] Specifically, the determination module is used to determine a target keyword sequence corresponding to the voice instruction when a voice instruction is detected; the target keyword sequence includes at least one target key word segmentation;

[0028] A voice module is used to search for a target audio corresponding to a target keyword sequence in a locally stored offline corpus, and to voice broadcast the found target audio; wherein the offline corpus includes each candidate keyword sequence and each candidate audio corresponding to each candidate keyword sequence, and each candidate keyword sequence includes at least one candidate key word.

[0029] Optionally, in a possible design, the voice broadcasting device provided by the present application may further include a sending module;

[0030] The determination module is further used to, after the speech module searches for the target audio corresponding to the target keyword sequence in the offline corpus stored locally, if the target audio corresponding to the target keyword sequence is not found in the offline corpus, determine whether the electrical device is in a networked state;

[0031] A sending module, used for sending a target keyword sequence or a voice command to a background server when the determination module determines that the electrical device is in a networked state, and for instructing the background server to determine a target audio according to the target keyword sequence or the voice command;

[0032] The voice module is also used to obtain the target audio returned by the background server, and to voice broadcast the target audio, and to update the corresponding relationship between the target keyword sequence and the target audio into the offline corpus.

[0033] Optionally, in another possible design, the sending module is specifically used for:

[0034] The target keyword sequence is sent to the backend server to instruct the backend server to search for the target audio corresponding to the target keyword sequence from the online corpus.

[0035] Optionally, in another possible design, the sending module is specifically used for:

[0036] Send a voice command to the background server to instruct the background server to process the voice command based on the preset voice recognition model and obtain the target audio output by the preset voice recognition model.

[0037] Optionally, in another possible design, the voice module is specifically used for:

[0038] Based on the target keyword sequence, determine the inquiry type to which the voice command belongs;

[0039] Based on the query type, a target query sub-corpus is determined from each query sub-corpus of the offline corpus;

[0040] Search the target audio corresponding to the target keyword sequence in the target query subcorpus.

[0041] Optionally, in another possible design, the inquiry type includes at least an activation inquiry class and a fault inquiry class.

[0042] Optionally, in another possible design, the voice module is specifically used for:

[0043] Analyze the voice command to determine the timbre characteristics of the voice command;

[0044] Determine the target user of the voice command based on the timbre characteristics;

[0045] Determine a target account sub-corpus corresponding to the target user from each account sub-corpus of the offline corpus;

[0046] Search the target audio corresponding to the target keyword sequence in the target account sub-corpus.

[0047] In a third aspect, the present application provides a voice broadcast device, comprising a memory, a processor, a bus and a communication interface; the memory is used to store computer execution instructions, and the processor is connected to the memory through a bus; when the voice broadcast device is running, the processor executes the computer execution instructions stored in the memory, so that the voice broadcast device executes the voice broadcast method provided in the first aspect above.

[0048] In a fourth aspect, the present application provides a computer-readable storage medium, in which instructions are stored. When a computer executes the instructions, the computer executes the voice broadcast method provided in the first aspect.

[0049] In a fifth aspect, the present application provides a computer program product, which includes computer instructions. When the computer instructions are executed on a computer, the computer executes the voice broadcast method provided in the first aspect.

[0050] It should be noted that the above computer instructions may be stored in whole or in part on a computer-readable storage medium, wherein the computer-readable storage medium may be packaged together with the processor of the voice broadcast device, or may be packaged separately from the processor of the voice broadcast device, which is not limited in this application.

[0051] The description of the second, third, fourth and fifth aspects of the present application can refer to the detailed description of the first aspect; and the beneficial effects of the description of the second, third, fourth and fifth aspects can refer to the beneficial effect analysis of the first aspect, which will not be repeated here.

[0052] In this application, the names of the above-mentioned devices or functional modules are not limited. In actual implementation, these devices or functional modules may appear with other names. As long as the functions of each device or functional module are similar to those of this application, they are within the scope of this application and its equivalent technologies.

[0053] These and other aspects of the present application will become more apparent from the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 A flowchart of a voice broadcast method provided in an embodiment of the present application;

[0055] Figure 2 A flowchart of another voice broadcast method provided in an embodiment of the present application;

[0056] Figure 3 A flowchart of another voice broadcast method provided in an embodiment of the present application;

[0057] Figure 4 A schematic diagram of the structure of a voice broadcast device provided in an embodiment of the present application;

[0058] Figure 5 A structural schematic diagram of a voice broadcast device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] The following is a detailed description of the voice broadcast method, device, equipment and storage medium provided in the embodiments of the present application in conjunction with the accompanying drawings.

[0060] The term "and / or" in this article is merely a description of the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone.

[0061] The terms "first" and "second" and the like in the specification and drawings of this application are used to distinguish different objects, or to distinguish different processing of the same object, rather than to describe a specific order of objects.

[0062] In addition, the terms "including" and "having" and any variations thereof mentioned in the description of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include other steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

[0063] It should be noted that, in the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0064] In the description of the present application, unless otherwise specified, “plurality” means two or more.

[0065] In addition, the acquisition, storage, use, and processing of data in the technical solution of this application comply with the relevant provisions of national laws and regulations.

[0066] In the prior art, when a user performs human-computer voice interaction with an electrical device, the electrical device must be connected to the Internet. When the electrical device is disconnected from the Internet, the user will not be able to use the voice interaction function, which affects the user experience.

[0067] In response to the problems existing in the above-mentioned prior art, the embodiments of the present application provide a voice broadcast method, device, equipment and storage medium. In this method, the human-computer voice interaction between the user and the electrical device is not restricted by the network status of the electrical device. When the electrical device is disconnected from the network, the user can still use the voice interaction function. Therefore, the embodiments of the present application can improve the user experience.

[0068] The voice broadcast method provided in the embodiment of the present application can be executed by the voice broadcast device provided in the embodiment of the present application, which can be implemented by software and / or hardware and integrated in the voice broadcast device that executes the method. The voice broadcast device can be an electrical device or a component in the electrical device, for example, it can be the main control board of the electrical device. For example, the electrical device can be a washing device.

[0069] The voice broadcast method provided in the embodiment of the present application is described below with reference to the accompanying drawings.

[0070] Reference Figure 1 The voice broadcast method provided in the embodiment of the present application includes S101-S102:

[0071] S101. When a voice command is detected, determine a target keyword sequence corresponding to the voice command.

[0072] The voice announcement method provided in the embodiment of the present application can be applied to electrical appliances, and for example, the electrical appliances can be washing appliances such as washing machines and clothes dryers. In the following description of the embodiment of the present application, the application of the voice announcement method to washing appliances will be described as an example.

[0073] In a possible implementation, the voice command may be a command detected in response to a voice command issued by a user to the electrical device. For example, the voice command issued by the user may be "Precautions for washing woolen clothes".

[0074] The target keyword sequence may be a sequence including at least one target keyword segmented word determined by performing voice recognition processing on a voice command corresponding to a voice instruction.

[0075] For example, taking the voice command "Precautions for washing woolen clothes" as an example, by performing voice recognition processing on the voice command, the three key participles "washing", "wool", and "precautions" can be obtained. By splicing these three target key participles, the target keyword sequence "washing; wool; precautions" can be obtained. In one possible implementation, the voice command can be processed for voice recognition using a pre-obtained word segmentation tool to obtain a target keyword sequence; in another possible implementation, the voice command can be processed for voice recognition using a pre-determined keyword recognition algorithm to obtain a target keyword sequence. Of course, in actual applications, voice recognition processing can also be performed on the voice command to determine the target keyword sequence in other ways, and the embodiments of the present application are not limited to this.

[0076] Optionally, before performing speech recognition processing on the voice command to obtain the target keyword sequence, the embodiment of the present application may also perform noise reduction processing on the voice command to filter out the ambient sound in the voice command and enhance the human voice in the voice command. In this way, the accuracy of the determined target keyword sequence can be improved.

[0077] S102: Searching for a target audio corresponding to the target keyword sequence in a locally stored offline corpus, and performing voice broadcasting of the found target audio.

[0078] The offline corpus may be a corpus that is determined in advance and stored locally. The offline corpus includes candidate keyword sequences and candidate audios corresponding to each candidate keyword sequence, and each candidate keyword sequence includes at least one candidate key word segmentation.

[0079] In a possible implementation, an offline corpus can be determined in advance based on the commonly used voice commands of most users, and the offline corpus is stored in the local memory of the electrical device when the electrical device leaves the factory. Each candidate keyword sequence in the offline corpus provided in the embodiment of the present application can cover the commonly used voice commands of most users, that is, it can meet the human-computer voice interaction needs of most users.

[0080] Exemplarily, the offline corpus includes candidate audio corresponding to the candidate keyword sequence of "washing; wool; precautions". When the target keyword sequence is determined to be "washing; wool; precautions", the target keyword sequence can be matched with each candidate keyword sequence in a predetermined order. When the three target key participles in the target keyword sequence "washing; wool; precautions" and the three candidate key participles in the candidate keyword sequence "washing; wool; precautions" are successfully matched, it can be determined that the target keyword sequence and the candidate keyword sequence are successfully matched, and the candidate audio corresponding to the successfully matched candidate keyword sequence can be determined as the target audio.

[0081] Optionally, after searching for the target audio corresponding to the target keyword sequence in the locally stored offline corpus, the voice broadcast method provided in the embodiment of the present application may also include: if the target audio corresponding to the target keyword sequence is not found in the offline corpus, determining whether the electrical device is in an online state; when it is determined that the electrical device is in an online state, sending the target keyword sequence or voice instruction to the background server to instruct the background server to determine the target audio based on the target keyword sequence or voice instruction; obtaining the target audio returned by the background server, voice broadcasting the target audio, and updating the correspondence between the target keyword sequence and the target audio to the offline corpus.

[0082] Since the offline corpus is fixedly stored in the memory of the electrical device when the electrical device leaves the factory, and the memory of the electrical device is limited, the voice interaction content that the offline corpus can cover is also limited. In this way, it may appear that the target audio corresponding to the target keyword sequence cannot be found in the offline corpus. Based on this, in the embodiment of the present application, when the target audio corresponding to the target keyword sequence is not found in the offline corpus, the target audio can be obtained online, and the target audio obtained online can be updated to the offline corpus synchronously, so that when the user issues the same voice command next time, if the electrical device is in a disconnected state, the corresponding target audio can be obtained from the offline corpus. It can be seen that in the technical solution provided by the embodiment of the present application, the electrical device can have both the offline voice broadcast function and the online voice broadcast function, and the user can use the online voice broadcast function when the electrical device is connected to the network (the offline voice broadcast function is preferred when connected to the network, that is, the target audio is first obtained from the offline corpus), and the offline voice broadcast function is used when the electrical device is disconnected. In addition, when using the online voice broadcast function, the target audio obtained during the current user's voice interaction can be stored to achieve dynamic expansion of the offline voice library. Then, as the user's use of electrical devices increases, the content covered in the offline voice library will become closer and closer to the current user's usage habits, thereby meeting the current user's personalized voice interaction needs and further improving the user experience.

[0083] In a possible implementation, the device corresponding to the background server may be a background server of an electrical device, and a communication module is provided in the electrical device, and the electrical device may send a target keyword sequence or a voice command to the background server through the communication module. Exemplarily, the communication module may be a wireless communication module such as a Bluetooth module, a ZigBee technology module, a 4G communication module, or a 5G communication module, and the embodiments of the present application do not limit this.

[0084] Optionally, a target keyword sequence or voice instruction is sent to a background server to instruct the background server to determine a target audio based on the target keyword sequence or voice instruction, including: sending a target keyword sequence to a background server to instruct the background server to search for a target audio corresponding to the target keyword sequence from an online corpus.

[0085] In a possible implementation, the backend server stores an online corpus, which may include device sub-expectation libraries corresponding to different electrical devices. When the voice broadcast device sends the target keyword sequence to the backend server, it may carry the device identification of the electrical device. The backend server may determine the target device sub-expectation library corresponding to the device identification from the online corpus based on the device identification, and then search for the target audio corresponding to the target keyword sequence in the target device sub-expectation library.

[0086] Among them, the online corpus contains a large number of candidate keyword sequences and corresponding candidate audios. Compared with the offline corpus, it has a wider coverage, and the content in the online corpus can also be updated regularly.

[0087] Optionally, sending a target keyword sequence or voice instruction to the background server to instruct the background server to determine the target audio based on the target keyword sequence or voice instruction may include: sending a voice instruction to the background server to instruct the background server to process the voice instruction based on a preset voice recognition model and obtain the target audio output by the preset voice recognition model.

[0088] The preset speech recognition model may be a preset speech recognition model pre-trained based on a training set (including a large number of sample speech commands and sample audio corresponding to each sample speech command) using a neural network or deep recognition algorithm. Exemplarily, the preset speech recognition model may be a generative pre-trained transformer (GPT) model based on a transformer.

[0089] Optionally, searching for target audio corresponding to the target keyword sequence in a locally stored offline corpus includes: determining the inquiry type to which the voice instruction belongs based on the target keyword sequence; determining a target inquiry sub-corpus from each inquiry sub-corpus of the offline corpus based on the inquiry type; and searching for target audio corresponding to the target keyword sequence in the target inquiry sub-corpus.

[0090] In order to improve the rate of successfully matching the target keyword sequence with the candidate keyword sequence in the offline corpus, thereby improving the rate of determining the target audio, in an embodiment of the present application, the voice instructions can be divided based on the query type, and the offline corpus can be divided into multiple query sub-corpora accordingly. When determining the target audio, the target query sub-corpus corresponding to the query type to which the voice instruction belongs can be first determined from each query sub-corpus, and then the target keyword sequence can be matched with the candidate keyword sequence in the target query sub-corpus to improve the matching efficiency.

[0091] Optionally, the inquiry type includes at least an activation inquiry class and a fault inquiry class.

[0092] Among them, the content in the inquiry sub-corpus corresponding to the enabled inquiry class may be voice interaction content related to the process of using the washing device, and the content in the inquiry sub-corpus corresponding to the fault inquiry class may be voice interaction content related to fault handling of the washing device.

[0093] For example, the voice command issued by the user may be "the amount of detergent required to wash 5 pieces of clothing", and the target keyword sequence corresponding to the voice command may be "washing, 5 pieces of clothing, detergent", and correspondingly, the candidate keyword sequence "washing, 5 pieces of clothing, detergent" and the corresponding candidate audio "about 20 ml" may be stored in the offline corpus. In this example, the inquiry type to which the voice command belongs is the enabling inquiry type.

[0094] In another example, the voice command issued by the user may be "The washing machine takes too long to drain", and the target keyword sequence corresponding to the voice command may be "drainage, time, long", and the candidate keyword sequence "drainage, time, long" and the corresponding candidate audio "Please check whether there is a blockage at the drainage. If there is a blockage, you can clear it yourself or contact the staff for on-site treatment" can be stored in the offline corpus. In this example, the inquiry type to which the voice command belongs is the fault inquiry type.

[0095] Optionally, a target audio corresponding to a target keyword sequence is searched in a locally stored offline corpus, including: parsing the voice command to determine the timbre characteristics of the voice command; determining the target user to whom the voice command belongs based on the timbre characteristics; determining a target account sub-corpus corresponding to the target user from each account sub-corpus of the offline corpus; and searching for the target audio corresponding to the target keyword sequence in the target account sub-corpus.

[0096] For the same electrical device, the user may have multiple users. In order to further meet the diverse needs of users, in the embodiment of the present application, a corresponding account sub-corpus can be created for each user. When determining the target audio, the target account sub-corpus corresponding to the target user can be first determined from each account sub-corpus of the offline corpus, and then the target keyword sequence can be matched with the candidate keyword sequence in the target account sub-corpus. In this way, the matching efficiency can be further improved, and the personalized needs of users can be further met.

[0097] Optionally, in practical applications, the query type and timbre features may be combined to match the target keyword sequence with the candidate keyword sequence in the offline corpus to further improve the matching efficiency.

[0098] Further optionally, taking the electrical appliance being a washing appliance as an example, searching for the target audio corresponding to the target keyword sequence in the locally stored offline corpus may also include: determining whether each target key participle of the target keyword sequence includes a clothing material key participle; in the case where it is determined that each target key participle of the target keyword sequence includes a target key participle associated with clothing material, determining a target material sub-corpus from the offline corpus based on the target key participle associated with clothing material; and searching for the target audio corresponding to the target keyword sequence in the target material sub-corpus.

[0099] The voice broadcast method provided in the embodiment of the present application can be applied to washing equipment. When a user uses the washing equipment, he or she may have different interactive contents for clothes of different materials. Therefore, the embodiment of the present application can also match the target keyword sequence with the candidate keyword sequence in the offline corpus based on the material characteristics. In this way, the matching efficiency can be further improved.

[0100] Based on the above description, in the voice broadcast method provided by the embodiment of the present application, an offline corpus including each candidate keyword sequence and the candidate audio corresponding to each candidate keyword sequence can be pre-stored locally. When a voice command is detected, the target keyword sequence corresponding to the voice command can be determined first; thereafter, the target audio corresponding to the target keyword sequence can be searched in the offline corpus, and the found target audio can be voice broadcasted. It can be seen that the embodiment of the present application can realize human-computer voice interaction through the offline corpus pre-stored locally. Since the offline corpus is pre-stored locally and does not need to be obtained online, in the voice broadcast method provided by the embodiment of the present application, the human-computer voice interaction between the user and the electrical device is not limited by the network status of the electrical device. When the electrical device is disconnected from the network, the user can still use the voice interaction function. Therefore, the present application can improve the user experience.

[0101] Optional, such as Figure 2 As shown, the embodiment of the present application further provides a voice broadcast method, which may include S201-S205:

[0102] S201: When a voice command is detected, determine a target keyword sequence corresponding to the voice command.

[0103] S202: Determine whether a target audio corresponding to the target keyword sequence is found in a locally stored offline corpus.

[0104] If the target audio corresponding to the target keyword sequence is found in the locally stored offline corpus, step S203 is executed; if the target audio corresponding to the target keyword sequence is not found in the locally stored offline corpus, step S204 is executed.

[0105] S203: Voice broadcast the found target audio.

[0106] S204. When it is determined that the electrical device is in an online state, a target keyword sequence or a voice command is sent to a background server to instruct the background server to determine a target audio according to the target keyword sequence or the voice command.

[0107] S205: Obtain the target audio returned by the background server, perform voice broadcasting on the target audio, and update the correspondence between the target keyword sequence and the target audio into the offline corpus.

[0108] Optional, such as Figure 3 As shown, the embodiment of the present application further provides a voice broadcast method, which may include S301-S306:

[0109] S301: When a voice instruction is detected, determine a target keyword sequence corresponding to the voice instruction.

[0110] S302: Determine the query type to which the voice instruction belongs based on the target keyword sequence, and determine a target query sub-corpus from each query sub-corpus of the offline corpus based on the query type.

[0111] S303: Analyze the voice command, determine the timbre characteristics of the voice command, and determine the target user to which the voice command belongs based on the timbre characteristics.

[0112] S304: Determine a target account sub-corpus corresponding to the target user from the target query sub-corpus.

[0113] S305. Search the target account sub-corpus for the target audio corresponding to the target keyword sequence.

[0114] S306: Voice broadcast the found target audio.

[0115] like Figure 4 As shown, an embodiment of the present application further provides a voice broadcast device, which can be configured in an electrical device, and the device can include: a determination module 11 and a voice module 21.

[0116] The determination module 11 executes S101 in the above method embodiment, and the speech module 21 executes S102 in the above method embodiment.

[0117] Specifically, the determination module 11 is used to determine the target keyword sequence corresponding to the voice instruction when the voice instruction is detected; the target keyword sequence includes at least one target key word segmentation;

[0118] The voice module 21 is used to search for the target audio corresponding to the target keyword sequence in the locally stored offline corpus, and voice broadcast the found target audio; wherein the offline corpus includes each candidate keyword sequence and the candidate audio corresponding to each candidate keyword sequence, and each candidate keyword sequence includes at least one candidate key word.

[0119] Optionally, in a possible design, the voice broadcasting device provided by the present application may further include a sending module;

[0120] The determination module 11 is also used to determine whether the electrical device is in a networked state if the target audio corresponding to the target keyword sequence is not found in the offline corpus after the speech module 21 searches for the target audio corresponding to the target keyword sequence in the locally stored offline corpus;

[0121] A sending module, used for sending a target keyword sequence or a voice command to a background server when the determining module 11 determines that the electrical device is in a networked state, and for instructing the background server to determine a target audio according to the target keyword sequence or the voice command;

[0122] The voice module 21 is also used to obtain the target audio returned by the background server, and to voice broadcast the target audio, and to update the corresponding relationship between the target keyword sequence and the target audio into the offline corpus.

[0123] Optionally, in another possible design, the sending module is specifically used for:

[0124] The target keyword sequence is sent to the backend server to instruct the backend server to search for the target audio corresponding to the target keyword sequence from the online corpus.

[0125] Optionally, in another possible design, the sending module is specifically used for:

[0126] Send a voice command to the background server to instruct the background server to process the voice command based on the preset voice recognition model and obtain the target audio output by the preset voice recognition model.

[0127] Optionally, in another possible design, the voice module 21 is specifically used for:

[0128] Based on the target keyword sequence, determine the inquiry type to which the voice command belongs;

[0129] Based on the query type, a target query sub-corpus is determined from each query sub-corpus of the offline corpus;

[0130] Search the target audio corresponding to the target keyword sequence in the target query subcorpus.

[0131] Optionally, in another possible design, the inquiry type includes at least an activation inquiry class and a fault inquiry class.

[0132] Optionally, in another possible design, the voice module 21 is specifically used for:

[0133] Analyze the voice command to determine the timbre characteristics of the voice command;

[0134] Determine the target user of the voice command based on the timbre characteristics;

[0135] Determine a target account sub-corpus corresponding to the target user from each account sub-corpus of the offline corpus;

[0136] Search the target audio corresponding to the target keyword sequence in the target account sub-corpus.

[0137] Optionally, the voice broadcast device may further include a storage module, and the storage module is used to store program codes of the voice broadcast device, etc.

[0138] like Figure 5 As shown, the embodiment of the present application also provides a voice broadcast device, which can be an electrical device or a component of an electrical device, such as a main control board of the electrical device. The voice broadcast device includes a memory 41, a processor (such as Figure 5 42-1 and 42-2), bus 43 and communication interface 44; memory 41 is used to store computer execution instructions, and the processor is connected to the memory 41 through bus 43; when the voice broadcast device is running, the processor executes the computer execution instructions stored in the memory 41, so that the voice broadcast device executes the voice broadcast method provided in the above embodiment.

[0139] In a specific implementation, as an embodiment, the processor may include one or more central processing units (CPUs), such as Figure 5 As an embodiment, the voice broadcast device may include multiple processors, such as Figure 5 4 and 5. The processors 42-1 and 42-2 shown in FIG. Each of these processors may be a single-core processor (single-CPU) or a multi-core processor (multi-CPU). The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer program instructions).

[0140] The memory 41 may be a read-only memory (ROM) or other types of static storage devices that can store static information and instructions, a random access memory (RAM) or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory 41 may exist independently and be connected to the processor via a bus 43. The memory 41 may also be integrated with the processor.

[0141] In a specific implementation, the memory 41 is used to store the data in this application and the computer execution instructions corresponding to the software program for executing this application. The processor can perform various functions of the voice broadcast device by running or executing the software program stored in the memory 41 and calling the data stored in the memory 41.

[0142] The communication interface 44 uses any transceiver-like device for communicating with other devices or communication networks, such as a control system, a radio access network (RAN), a wireless local area network (WLAN), etc. The communication interface 44 may include a receiving unit to implement a receiving function, and a sending unit to implement a sending function.

[0143] The bus 43 may be an industry standard architecture (ISA) bus, a peripheral component interconnect (PCI) bus, or an extended industry standard architecture (EISA) bus. The bus 43 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0144] As an example, combining Figure 4 , the function of the determination module in the voice broadcast device is the same as Figure 5 When the voice broadcast device includes a sending module, the function implemented by the sending module is the same as Figure 5 The function implemented by the sending unit in the voice broadcast device is the same as that implemented by the sending unit in the voice broadcast device. When the voice broadcast device includes a storage module, the function implemented by the storage module is the same as that implemented by the sending unit in the voice broadcast device. Figure 5 The memory in implements the same function.

[0145] The explanation of the relevant contents in this embodiment can refer to the above method embodiment, which will not be repeated here.

[0146] Through the description of the above implementation methods, technicians in the relevant field can clearly understand that for the convenience and simplicity of description, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. The specific working process of the system, device and unit described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0147] The embodiment of the present application also provides a computer-readable storage medium, in which instructions are stored. When a computer executes the instructions, the computer executes the voice broadcast method provided in the above embodiment.

[0148] Among them, the computer-readable storage medium, for example, can be but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, RAM, ROM, an erasable programmable read only memory (EPROM), a register, a hard disk, an optical fiber, a CD-ROM, an optical storage device, a magnetic storage device, or any suitable combination of the above, or any other form of computer-readable storage medium known in the art. An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application-specific integrated circuit (ASIC). In an embodiment of the present application, the computer-readable storage medium can be any tangible medium containing or storing a program, which can be used by or in combination with an instruction execution system, device or device.

[0149] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A voice broadcast method, characterized in that: Applied to electrical equipment, including: In the case where a voice instruction is detected, determining a target keyword sequence corresponding to the voice instruction; the target keyword sequence includes at least one target key word; Search for target audio corresponding to the target keyword sequence in a locally stored offline corpus, and voice broadcast the found target audio; wherein the offline corpus includes each candidate keyword sequence and candidate audio corresponding to each candidate keyword sequence, and each candidate keyword sequence includes at least one candidate key word.

2. The voice broadcast method according to claim 1, characterized in that: After searching the locally stored offline corpus for the target audio corresponding to the target keyword sequence, the method further includes: If the target audio corresponding to the target keyword sequence is not found in the offline corpus, determining whether the electrical device is in a networked state; When it is determined that the electrical device is in the networking state, sending the target keyword sequence or the voice instruction to the background server to instruct the background server to determine the target audio according to the target keyword sequence or the voice instruction; The target audio returned by the background server is obtained, and the target audio is voice broadcasted, and the corresponding relationship between the target keyword sequence and the target audio is updated in the offline corpus.

3. The voice broadcast method according to claim 2, characterized in that: The sending the target keyword sequence or the voice instruction to the background server to instruct the background server to determine the target audio according to the target keyword sequence or the voice instruction includes: The target keyword sequence is sent to the background server to instruct the background server to search the target audio corresponding to the target keyword sequence from an online corpus.

4. The voice broadcast method according to claim 2, characterized in that: The sending the target keyword sequence or the voice instruction to the background server to instruct the background server to determine the target audio according to the target keyword sequence or the voice instruction includes: The voice instruction is sent to the background server to instruct the background server to process the voice instruction based on a preset voice recognition model and obtain the target audio output by the preset voice recognition model.

5. The voice broadcast method according to claim 1, characterized in that: The step of searching for a target audio corresponding to the target keyword sequence in a locally stored offline corpus includes: Determining the inquiry type to which the voice instruction belongs based on the target keyword sequence; Based on the query type, determining a target query sub-corpus from each query sub-corpus of the offline corpus; The target audio corresponding to the target keyword sequence is searched in the target query sub-corpus.

6. The voice broadcast method according to claim 5, characterized in that: The inquiry types include at least an activation inquiry type and a fault inquiry type.

7. The voice broadcast method according to claim 1, characterized in that: The step of searching for a target audio corresponding to the target keyword sequence in a locally stored offline corpus includes: Analyzing the voice command to determine the timbre characteristics of the voice command; Based on the timbre feature, determining a target user to which the voice command belongs; Determining a target account sub-corpus corresponding to the target user from each account sub-corpus of the offline corpus; The target audio corresponding to the target keyword sequence is searched in the target account sub-corpus.

8. A voice broadcasting device, characterized in that: Configured in electrical equipment, including: A determination module, configured to determine, when a voice instruction is detected, a target keyword sequence corresponding to the voice instruction; the target keyword sequence includes at least one target key word segmentation; A voice module is used to search for a target audio corresponding to the target keyword sequence in a locally stored offline corpus, and to voice broadcast the found target audio; wherein the offline corpus includes each candidate keyword sequence and the candidate audio corresponding to each candidate keyword sequence, and each candidate keyword sequence includes at least one candidate key word.

9. A voice broadcasting device, characterized in that: It includes a memory, a processor, a bus and a communication interface; the memory is used to store computer-executable instructions, and the processor is connected to the memory via the bus; When the voice broadcast device is running, the processor executes the computer-executable instructions stored in the memory, so that the voice broadcast device executes the voice broadcast method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, and when the computer executes the instructions, the computer executes the voice broadcast method as described in any one of claims 1-7.

Images