Sentence disassembly method, device and equipment based on optional items, and storage medium

Abstract

The application discloses a sentence disassembly method and device based on optional items, equipment and a storage medium, and belongs to the technical field of speech recognition. The application obtains an input voice data set, determines optional items and mandatory items in the voice data set, adds corresponding identifiers to the voice data set according to the optional items and the mandatory items, and performs sentence disassembly on the voice data set based on the identifiers. The above-mentioned method can meet the requirements of sentence disassembly containing optional items, and greatly improves the disassembly efficiency when disassembling sentences containing optional items in the voice.

Description

Technical Field

[0001] This invention relates to the field of speech recognition technology, and in particular to a method, apparatus, device, and storage medium for sentence decomposition based on optional options. Background Technology

[0002] The current methods for sorting permutations of speech statements can only sort permutations that are all mandatory options, and cannot sort permutations that contain optional options. Therefore, the current decomposition algorithm cannot meet the requirements for decomposing statements with optional options, which greatly reduces the efficiency of statement decomposition.

[0003] The above content is only used to help understand the technical solution of the present invention and does not represent an admission that the above content is prior art. Summary of the Invention

[0004] The main objective of this invention is to provide a method, apparatus, device, and storage medium for decomposing statements based on optional options, aiming to solve the technical problem that the existing technology cannot meet the requirements for decomposing statements containing optional options, which greatly reduces the efficiency of statement decomposition.

[0005] To achieve the above objectives, the present invention provides a statement decomposition method based on optionalities, the statement decomposition method based on optionalities comprising the following steps:

[0006] Obtain the input speech dataset;

[0007] Determine the optional and required options in the speech dataset;

[0008] Add corresponding identifiers to the speech dataset based on the optional and required options;

[0009] The speech dataset is decomposed based on the identifier.

[0010] Optionally, determining the optional and required options in the speech dataset includes:

[0011] The speech dataset is split into multiple words, and the multiple words are arranged and combined to obtain the speech sentences;

[0012] Obtain the identifiers of each word in the statement;

[0013] The optional and required options in the speech dataset are determined based on the identifiers of each word.

[0014] Optionally, adding corresponding identifiers to the speech dataset based on the optional and required options includes:

[0015] Add an identifier to the optional option to convert it into a required option, while keeping the required option unchanged.

[0016] Optionally, the step of decomposing the speech dataset based on the identifier includes:

[0017] Identify the required options in the speech dataset that contain the identifier;

[0018] The speech dataset is decomposed once based on the required options containing the identifier to obtain a reference combination of required options;

[0019] The reference mandatory combination is further decomposed to obtain the target mandatory combination.

[0020] Optionally, the step of decomposing the speech dataset according to the required options containing the identifier to obtain a reference combination of required options includes:

[0021] Identify the identifier corresponding to the required option containing the identifier and the converted required option, wherein the required option is obtained by converting the optional option;

[0022] Choose one of the identifier and the converted required options and combine it with a required option that does not contain the identifier to obtain a reference required option combination.

[0023] Optionally, the further decomposition of the reference mandatory option combination to obtain the target mandatory option combination includes:

[0024] Remove the identifiers from the reference mandatory combination to obtain the remaining mandatory combination;

[0025] The remaining mandatory option combinations are further decomposed to obtain the target mandatory option combinations.

[0026] Optionally, the optional words are words that can be omitted from the statement, and the required words are words that cannot be omitted from the statement.

[0027] Furthermore, to achieve the above objectives, the present invention also proposes a statement decomposition device based on optionalities, the statement decomposition device based on optionalities comprising:

[0028] The acquisition module is used to acquire the input speech dataset;

[0029] The recognition module is used to determine the optional and required options in the speech dataset;

[0030] The processing module is used to add corresponding identifiers to the voice dataset according to the optional and the required options;

[0031] The processing module is also used to perform sentence decomposition on the speech dataset based on the identifier.

[0032] Furthermore, to achieve the above objectives, the present invention also proposes an option-based statement decomposition device, which includes: a memory, a processor, and an option-based statement decomposition program stored in the memory and running on the processor, wherein the option-based statement decomposition program is configured to implement the option-based statement decomposition method as described above.

[0033] Furthermore, to achieve the above objectives, the present invention also proposes a storage medium storing an option-based statement decomposition program, which, when executed by a processor, implements the option-based statement decomposition method as described above.

[0034] This invention acquires an input speech dataset; determines optional and required options in the speech dataset; adds corresponding identifiers to the speech dataset based on the optional and required options; and performs sentence decomposition on the speech dataset based on the identifiers. By first identifying optional and required options in the speech dataset, then adding identifiers based on the identified optional and required options, and finally performing final sentence decomposition based on the added identifiers, this method can meet the requirements for decomposing sentences containing optional options, and greatly improves the decomposition efficiency when decomposing sentences containing optional options in speech. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the structure of the hardware operating environment based on optional statement disassembly device involved in the embodiments of the present invention;

[0036] Figure 2 This is a flowchart illustrating the first embodiment of the optional statement decomposition method of the present invention;

[0037] Figure 3 This is a schematic diagram of the statement decomposition process in one embodiment of the optional statement decomposition method of the present invention;

[0038] Figure 4 This is a flowchart illustrating the second embodiment of the optional statement decomposition method of the present invention;

[0039] Figure 5 This is a flowchart illustrating the third embodiment of the optional statement decomposition method of the present invention;

[0040] Figure 6 This is a structural block diagram of the first embodiment of the optional statement decomposition device of the present invention.

[0041] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0042] It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of the invention.

[0043] Reference Figure 1 , Figure 1 This is a schematic diagram of the hardware operating environment based on optional statement disassembly of the device structure involved in the embodiments of the present invention.

[0044] like Figure 1 As shown, the optional statement splitting device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wireless-Fidelity (Wi-Fi) interface). The memory 1005 may be a high-speed random access memory (RAM) or a stable non-volatile memory (NVM), such as a disk drive. The memory 1005 may also optionally be a storage device independent of the aforementioned processor 1001.

[0045] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on the device based on optional statements, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0046] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a network communication module, a user interface module, and an optional statement decomposition program.

[0047] exist Figure 1In the optional statement decomposition device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and memory 1005 in the optional statement decomposition device of the present invention can be set in the optional statement decomposition device, and the optional statement decomposition device calls the optional statement decomposition program stored in the memory 1005 through the processor 1001 and executes the optional statement decomposition method provided in the embodiment of the present invention.

[0048] This invention provides a method for decomposing statements based on optional parameters, referring to... Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of a statement decomposition method based on optional options according to the present invention.

[0049] In this embodiment, the statement decomposition method based on optional options includes the following steps:

[0050] Step S10: Obtain the input speech dataset.

[0051] In this embodiment, the execution entity is the optional statement decomposition device, which has functions such as data acquisition, data communication, and program execution. Of course, other devices with similar functions can also be used; this embodiment does not limit this. This embodiment uses an optional statement decomposition device as an example for explanation.

[0052] It's important to note that for voice commands used in vehicles, users may express themselves in many different ways, including some redundant or optional content. However, the voice command model stored in the system during development is fixed and contains no extraneous information. For example, different users may use multiple different ways to express the same operation. Regardless of the expression, each contains both necessary and redundant content. If redundant content were not allowed, users would be forced to express themselves according to a set set of methods, significantly reducing the user experience. Therefore, when a command is received, the mandatory content needs to be identified so that it can be compared with the command model to clarify the specific semantics of the voice.

[0053] However, the current method can only sort permutations and combinations that are all mandatory options, and cannot sort permutations and combinations that contain optional options. In other words, it cannot decompose optional options. Therefore, the current decomposition algorithm cannot meet the requirements of decomposing statements containing optional options, which greatly reduces the efficiency of statement decomposition.

[0054] In this embodiment, to solve the aforementioned technical problem, the following steps are taken: First, an input speech dataset is acquired; then, optional and required options in the speech dataset are determined; corresponding identifiers are added to the speech dataset based on the optional and required options; finally, the speech dataset is decomposed based on the identifiers. This method first identifies the optional and required options in the speech dataset, then adds identifiers based on the identified optional and required options, and finally performs the final sentence decomposition based on the added identifiers. This approach satisfies the requirements for decomposing sentences containing optional options and greatly improves the decomposition efficiency when decomposing speech sentences containing optional options. Specifically, it can be implemented as follows.

[0055] In this embodiment, first according to Figure 3 Taking an example, the overall process of this technical solution will be illustrated. (Refer to...) Figure 3 First, the input speech dataset needs to be obtained. The obtained speech dataset consists of [I want|I want] (close|turn off) [voice|voice assistant]. Here, "I want" and "I want", "close" and "turn off", and "voice" and "voice assistant" are all different ways of expressing the same meaning. After obtaining the above speech dataset, it is necessary to further identify the optional and required options in the speech dataset. The optional options in the above speech dataset are [I want|I want] and [voice|voice assistant], and the required option is (close|turn off).

[0056] After identifying the optional and required options, an identifier is added to each optional option. The identifier can be set to Target. After adding the identifier Target, the voice dataset becomes ([I want|I want]Target)(Close|Turn off)([Voice|Voice Assistant]Target). Since the identifier has been added, the optional options are now converted into required options, for example ((I want|I want)Target)(Close|Turn off)((Voice|Voice Assistant)Target).

[0057] After adding the identifiers, the newly obtained speech dataset is first decomposed to obtain multiple mandatory combinations containing "Target". For example, combination one is (I want|I want)(Close|Turn off)(Voice|Voice Assistant), combination two is (I want|I want)(Close|Turn off)Target, combination three is Target(Close|Turn off)(Voice|Voice Assistant), and combination four is Target(Close|Turn off)Target. The above method is the Cartesian decomposition algorithm. The Cartesian product, also known as the direct product, is the Cartesian product of two sets X and Y in mathematics, represented as X×Y. The first object is a member of X, and the second object is one of the members of all possible ordered pairs of Y. Of course, other methods can be used to decompose the sentences in this embodiment, and there are no restrictions on this.

[0058] After obtaining the above combinations from the first statement decomposition, further decomposition is required. Before this second decomposition, the setting identifier Target needs to be removed. After removing Target, the resulting combinations are: Combination 1: (I want | I want) (Close | Turn off) (Voice | Voice Assistant); Combination 2: (I want | I want) (Close | Turn off); Combination 3: (Close | Turn off) (Voice | Voice Assistant); Combination 4: (Close | Turn off). Finally, a second decomposition of these combinations yields the final multiple mandatory combinations, such as "I want to close the voice," "I want to close the voice assistant," ..., "Turn off." See the attached document for details. Figure 3 As shown, further details will not be elaborated here. It is important to emphasize that all the statements obtained after decomposition are different expressions of the same statement. For example, expressions such as "I want to turn off the voice," "I want to turn off the voice assistant," ..., and "turn off" all correspond to the same meaning: turning off the voice or voice assistant.

[0059] In this specific implementation, it is necessary to first obtain a speech dataset. This speech dataset is a speech dataset that has not been processed by sentence decomposition and cannot be directly used in subsequent speech recognition to compare with the instruction model sentences. This speech dataset can be input by the user or obtained through other means. This embodiment does not impose any restrictions on this.

[0060] Step S20: Determine the optional and required options in the speech dataset.

[0061] In specific implementation, after obtaining the input speech dataset, this embodiment needs to further identify the optional and required options in the speech dataset. In this embodiment, optional options are words that can be omitted in the sentence, and required options are words that cannot be omitted in the sentence. If a required option is missing, the specific meaning of the sentence cannot be clearly defined, while if an optional option is missing, it will not affect the final semantics of the entire sentence.

[0062] It should be noted that the optional and required options in this embodiment may contain one or more words, and the specific number of words can be determined by the speech dataset, as described above. Figure 3 For example, the optional [I want|I think] options include "I want" and "I think", while the required options (close|turn off) include "close" and "turn off".

[0063] Step S30: Add corresponding identifiers to the voice dataset according to the optional and required options.

[0064] In specific implementation, after identifying the optional and required options in this embodiment, it is also necessary to add identifiers. Specifically, in this embodiment, an identifier can be added to each optional option, and the added identifier can be as follows: Figure 3 The Target shown can, of course, be set to other forms of identifiers according to actual needs, but this embodiment does not impose any restrictions on this.

[0065] Step S40: Decompose the speech dataset into sentences based on the identifier.

[0066] In this specific implementation, after adding the identifier to the speech dataset, the speech dataset is decomposed into sentences based on the identifier.

[0067] Specifically, the statement decomposition in this embodiment can be divided into two decomposition processes. The first decomposition process is to decompose the required options containing identifiers. The second decomposition process is to remove the identifiers and finally decompose all the required options. The decomposition can be performed by permutation and combination using Cartesian product, or other methods can be used. This embodiment does not impose any restrictions on this.

[0068] It should be understood that the final combination obtained from the breakdown is a set of mandatory options. This means that users must input their voice according to the multiple expressions corresponding to these mandatory options during subsequent voice control. For example, if a user wants to turn off the voice assistant, they must select the option that corresponds to their voice input. Figure 3 The 18 final expressions shown are input into the corresponding voice.

[0069] This embodiment acquires an input speech dataset; determines the optional and required options in the speech dataset; adds corresponding identifiers to the speech dataset based on the optional and required options; and performs sentence decomposition on the speech dataset based on the identifiers. By first identifying the optional and required options in the speech dataset, then adding identifiers based on the identified optional and required options, and finally performing the final sentence decomposition based on the added identifiers, the above method can meet the requirements for decomposing sentences containing optional options, and greatly improves the decomposition efficiency when decomposing sentences containing optional options in speech.

[0070] refer to Figure 4 , Figure 4 This is a flowchart illustrating a second embodiment of the optional statement decomposition method of the present invention.

[0071] Based on the first embodiment described above, a second embodiment of the present invention is proposed, which is a statement decomposition method based on optional options.

[0072] In this embodiment, in the optional statement decomposition method, step S20 specifically includes:

[0073] Step S201: Split the speech dataset into multiple words, and arrange and combine the multiple words to obtain the speech sentences.

[0074] In practice, the speech dataset consists of several words. After obtaining the input speech dataset, this embodiment can first split the speech dataset into multiple words, and then arrange and combine the resulting words to obtain the speech sentence. For example, if the speech dataset contains the words "I want", "close", and "voice assistant", then after arranging and combining them, the complete sentence "I want to close the voice assistant" can be obtained.

[0075] Step S202: Obtain the identifiers of each word in the statement.

[0076] In specific implementation, before identifying optional and required options in this embodiment, it is necessary to first obtain the identifiers corresponding to each word in the statement. The identifiers are used for subsequent identification of optional and required options.

[0077] Step S203: Determine the optional and required options in the speech dataset based on the identifiers of each word.

[0078] In practice, after obtaining the identifiers of each word, the optional and required options in the speech dataset can be determined based on these identifiers. Specifically, in this embodiment, the identifiers of each word can be compared with preset identifiers, which can be pre-set based on the required required options. Through the above comparison, the optional and required options in the speech dataset can be determined.

[0079] Further, step S30 specifically includes:

[0080] Step S301: Add an identifier to the optional option to convert the optional option into a required option, and keep the required option unchanged.

[0081] In specific implementation, after identifying optional and required options, the process of adding identifiers in this embodiment involves adding identifiers to optional options to convert them into required options. For example, if an optional option is [I want|I think], after adding the identifier Target, the optional option becomes a required option, such as (I want|I think)Target. Furthermore, if it is a required option, this embodiment does not add identifiers to the required option, and the required option remains unchanged.

[0082] This embodiment splits the speech dataset into multiple words and arranges and combines these words to obtain speech sentences; it obtains the identifiers of each word in the sentences; and it determines the optional and required options in the speech dataset based on the identifiers of each word. This accurately identifies the optional and required options in the sentences, and adds identifiers to the optional options to convert them into required options while keeping the required options unchanged. This facilitates subsequent sentence decomposition and improves decomposition efficiency.

[0083] refer to Figure 5 , Figure 5 This is a flowchart illustrating a third embodiment of a statement decomposition method based on optional options according to the present invention.

[0084] Based on the first embodiment described above, a third embodiment of the present invention is proposed, which is a statement decomposition method based on optional options.

[0085] In this embodiment, in the optional statement decomposition method, step S40 specifically includes:

[0086] Step S401: Identify the required options in the speech dataset that contain the identifier.

[0087] In practice, before disassembling, it is necessary to identify the required options containing identifiers, such as ((I want|I want)Target)(Close|Turn off)((Voice|Voice assistant)Target), where (I want|I want)Target is a required option containing an identifier.

[0088] Step S402: Decompose the speech dataset once according to the required options containing the identifier to obtain a reference required option combination.

[0089] In specific implementation, this embodiment requires a first decomposition of the required options containing identifiers. Specifically, it can identify the identifiers and converted required options corresponding to the required options containing identifiers. Then, it can randomly select one of the identifiers and converted required options to combine with the required options without identifiers. That is, the identifier is decomposed as a single object. For example, (I want|I want)Target and Target are decomposed as separate objects. For example, ((I want|I want)Target)(Close|Turn off)((Voice|Voice Assistant)Target) can be decomposed. The reference required option combinations that can be obtained include (I want|I want)(Close|Turn off)(Voice|Voice Assistant), (I want|I want)(Close|Turn off)Target, Target(Close|Turn off)(Voice|Voice Assistant), and Target(Close|Turn off)Target.

[0090] Step S403: Perform a second decomposition on the reference mandatory option combination to obtain the target mandatory option combination.

[0091] In specific implementation, after obtaining the above-mentioned reference mandatory option combinations, this embodiment needs to perform a secondary decomposition of the reference mandatory option combinations. Specifically, the process can be to first remove the identifier from the reference mandatory option combinations to obtain the remaining mandatory option combinations. For example, after removing the identifier Target from the above-mentioned reference mandatory option combinations, the remaining mandatory option combinations are (I want|I want)(Close|Turn off)(Voice|Voice Assistant), (I want|I want)(Close|Turn off), (Close|Turn off)(Voice|Voice Assistant), and (Close|Turn off). Then, the remaining mandatory option combinations are further decomposed to obtain the target mandatory option combinations. The target mandatory option combinations obtained by performing a secondary decomposition of the reference mandatory combinations are, for example, "I want to turn off voice," "I want to turn off voice assistant," ..., "Turn off," etc.

[0092] This embodiment identifies the required options containing the identifier in the speech dataset, performs a first decomposition of the speech dataset based on the required options containing the identifier to obtain a reference required option combination, and performs a second decomposition of the reference required option combination to obtain a target required option combination. The above method can meet the requirements for decomposing sentences containing optional options, and greatly improves the decomposition efficiency when decomposing sentences containing optional options in speech.

[0093] Furthermore, embodiments of the present invention also propose a storage medium storing an option-based statement decomposition program, wherein when the option-based statement decomposition program is executed by a processor, it implements the steps of the option-based statement decomposition method described above.

[0094] Since this storage medium adopts all the technical solutions of all the above embodiments, it has at least all the beneficial effects brought about by the technical solutions of the above embodiments, which will not be repeated here.

[0095] Reference Figure 6 , Figure 6 This is a structural block diagram of the first embodiment of the optional statement decomposition device of the present invention.

[0096] like Figure 6 As shown, the optional statement decomposition device proposed in this embodiment of the invention includes:

[0097] Module 10 is used to acquire the input speech dataset.

[0098] The recognition module 20 is used to determine the optional and required options in the speech dataset.

[0099] Processing module 30 is used to add corresponding identifiers to the voice dataset according to the optional and the required options.

[0100] The processing module 30 is further configured to perform sentence decomposition on the speech dataset based on the identifier.

[0101] This embodiment acquires an input speech dataset; determines the optional and required options in the speech dataset; adds corresponding identifiers to the speech dataset based on the optional and required options; and performs sentence decomposition on the speech dataset based on the identifiers. By first identifying the optional and required options in the speech dataset, then adding identifiers based on the identified optional and required options, and finally performing the final sentence decomposition based on the added identifiers, the above method can meet the requirements for decomposing sentences containing optional options, and greatly improves the decomposition efficiency when decomposing sentences containing optional options in speech.

[0102] In one embodiment, the recognition module 20 is further configured to split the speech dataset into multiple words, and arrange and combine the multiple words to obtain a speech sentence; obtain the identifier of each word in the sentence; and determine the optional and mandatory options in the speech dataset based on the identifier of each word.

[0103] In one embodiment, the processing module 30 is further configured to add an identifier to the optional option to convert the optional option into a required option, and keep the required option unchanged.

[0104] In one embodiment, the processing module 30 is further configured to identify the required options containing the identifier in the speech dataset; to perform a first decomposition of the speech dataset based on the required options containing the identifier to obtain a reference required option combination; and to perform a second decomposition of the reference required option combination to obtain a target required option combination.

[0105] In one embodiment, the processing module 30 is further configured to identify the identifier corresponding to the required option containing the identifier and the converted required option, wherein the required option is obtained by converting the optional option; and to randomly select one of the identifier and the converted required option and combine it with a required option that does not contain the identifier to obtain a reference required option combination.

[0106] In one embodiment, the processing module 30 is further configured to remove the identifier from the reference mandatory combination to obtain the remaining mandatory combination; and to perform a secondary decomposition on the remaining mandatory combination to obtain the target mandatory combination.

[0107] In one embodiment, the optional words are words that can be omitted from the statement, and the required words are words that cannot be omitted from the statement.

[0108] It should be understood that the above are merely illustrative examples and do not constitute any limitation on the technical solutions of the present invention. In specific applications, those skilled in the art can make settings as needed, and the present invention does not impose any restrictions on this.

[0109] It should be noted that the workflow described above is merely illustrative and does not limit the scope of protection of this invention. In practical applications, those skilled in the art can select some or all of the workflow to achieve the purpose of this embodiment according to actual needs, and no restrictions are imposed here.

[0110] In addition, for technical details not described in detail in this embodiment, please refer to the optional statement decomposition method provided in any embodiment of the present invention, which will not be repeated here.

[0111] Furthermore, it should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0112] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0113] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as read-only memory (ROM) / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0114] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A statement decomposition method based on optional options, characterized in that, The optional statement decomposition method includes: Obtain the input speech dataset; Determine the optional and required options in the speech dataset, where optional options are words that can be omitted from the sentence, and required options are words that cannot be omitted from the sentence; Add corresponding identifiers to the speech dataset based on the optional and required options; The speech dataset is decomposed based on the identifier; The step of decomposing the speech dataset based on the identifier includes: Identify the required options in the speech dataset that contain the identifier; The speech dataset is decomposed once based on the required options containing the identifier to obtain a reference combination of required options; The reference mandatory combination is further decomposed to obtain the target mandatory combination; The step of disassembling the speech dataset based on the required options containing the identifier to obtain a reference combination of required options includes: Identify the identifier corresponding to the required option containing the identifier and the converted required option, wherein the required option is obtained by converting the optional option; Choose one of the identifiers and the converted required options and combine it with a required option that does not contain the identifier to obtain a reference required option combination.

2. The statement decomposition method based on optionalities as described in claim 1, characterized in that, Determining the optional and required options in the speech dataset includes: The speech dataset is split into multiple words, and the multiple words are arranged and combined to obtain the speech sentences; Obtain the identifiers of each word in the statement; The optional and required options in the speech dataset are determined based on the identifiers of each word.

3. The statement decomposition method based on optionalities as described in claim 1, characterized in that, The step of adding corresponding identifiers to the speech dataset based on the optional and required options includes: Add an identifier to the optional option to convert it into a required option, while keeping the required option unchanged.

4. The statement decomposition method based on optionalities as described in claim 1, characterized in that, The process of further decomposing the reference mandatory option combination to obtain the target mandatory option combination includes: Remove the identifiers from the reference mandatory combination to obtain the remaining mandatory combination; The remaining mandatory option combinations are further decomposed to obtain the target mandatory option combinations.

5. A statement decomposition device based on optional elements, characterized in that, The optional statement decomposition device includes: The acquisition module is used to acquire the input speech dataset; The recognition module is used to determine the optional and required options in the speech dataset, wherein the optional options are words that can be omitted in the sentence, and the required options are words that cannot be omitted in the sentence; The processing module is used to add corresponding identifiers to the voice dataset according to the optional and the required options; The processing module is also used to perform sentence decomposition on the speech dataset based on the identifier; The processing module is further configured to identify the required options containing the identifier in the speech dataset; to perform a first decomposition of the speech dataset based on the required options containing the identifier to obtain a reference required option combination; and to perform a second decomposition of the reference required option combination to obtain a target required option combination. The processing module is further configured to identify the identifier corresponding to the required option containing the identifier and the converted required option, wherein the required option is obtained by converting the optional option; and to randomly select one of the identifier and the converted required option and combine it with the required option that does not contain the identifier to obtain a reference required option combination.

6. A statement decomposition device based on optional parameters, characterized in that, The optional statement splitting device includes: a memory, a processor, and an optional statement splitting program stored on the memory and running on the processor, the optional statement splitting program being configured to implement the optional statement splitting method as described in any one of claims 1 to 4.

7. A storage medium, characterized in that, The storage medium stores an option-based statement decomposition program, which, when executed by a processor, implements the option-based statement decomposition method as described in any one of claims 1 to 4.

Images