阅读说明：本技术 离线语义解析方法及系统 (Offline semantic parsing method and system ) 是由 陈家欢 甘津瑞 于 2020-07-21 设计创作，主要内容包括：本发明公开一种离线语义解析方法和系统,方法包括：根据多个语言字段获取与其对应的多个功能语义字段,在语言模型的gram语法编译内容中,对应多个功能语义字段创建多个语义分类字段skill_id；判断当前字段是否有对应多个语义分类字段skill_id,若是,则根据当前字段、语义对应关系获取对应的当前语义分类字段skill_id；若否,则根据当前字段及本地导航解析程序获取导航信息返回字段。本发明通过在gram语法编译的流程中添加skill id字段,为每一个说法都注明其分类,在给出识别结果的同时,给出当前说法属于哪一类的信息：skill id。通过skill id直接定位到对应类别的语义进行解析。(The invention discloses an off-line semantic analysis method and a system, wherein the method comprises the following steps: acquiring a plurality of corresponding functional semantic fields according to the plurality of language fields, and creating a plurality of semantic classification fields kill _ id corresponding to the plurality of functional semantic fields in the gram grammar compiling content of the language model; judging whether the current field has a plurality of corresponding semantic classification fields skill _ id, if so, acquiring the corresponding current semantic classification field skill _ id according to the corresponding relation between the current field and the semantic; if not, the navigation information return field is obtained according to the current field and the local navigation analysis program. The method adds the kill id field in the flow of compiling the gram grammar, notes the classification of each statement, and gives the information of which category the current statement belongs to while giving the recognition result: a kill id. And directly positioning the semantics of the corresponding category through the skip id for analysis.)

1. An offline semantic parsing method, comprising:

step S101, acquiring a plurality of corresponding functional semantic fields according to a plurality of language fields; establishing semantic corresponding relations between the plurality of functional semantic fields and the plurality of language fields;

step S102, in the gram grammar compiling content of the language model, a plurality of semantic classification fields kill _ id are created corresponding to the plurality of functional semantic fields; associating a plurality of semantic resolutions according to the plurality of semantic classification fields kill _ id; the plurality of semantic resolutions can obtain a determined output field;

step S103, judging whether the current field has a plurality of corresponding semantic classification fields kill _ id, if so, turning to step S104, and if not, turning to step S105;

step S104, obtaining a corresponding current semantic classification field kill _ id according to the current field and the semantic corresponding relation; analyzing and acquiring a current output field through corresponding semantics according to the current semantic classification field skill _ id;

step S105, navigation information return fields are obtained according to the current fields and a local navigation analysis program; the navigation analysis program can obtain corresponding navigation position data according to the current field.

2. The method of claim 1, wherein the step S103 comprises:

step S1031, acquiring a corresponding semantic classification field kill _ id according to the current field and the semantic corresponding relation;

step S1032, determining whether the semantic classification field kill _ id is empty, if not, going to step S104, and if yes, going to step S105.

3. The method of claim 1, wherein the step S104 comprises:

judging whether the current output field can be obtained through corresponding semantic analysis of the current semantic classification field kill _ id, and if so, outputting the current output field; if not, the process goes to step S105.

4. The method of claim 1 or 3, wherein the navigation parser corresponds to navigation parse semantics; the semantic corresponding relation comprises a corresponding relation between navigation analysis semantics and language fields;

the step S105 includes:

step S1051, obtaining the corresponding current functional semantic according to the current field;

step S1052, judging whether the current functional semantic is a navigation analysis semantic, if so, acquiring a navigation information return field according to the current field and a local navigation analysis program; and if not, identifying the current field through the natural language to obtain a current natural language identification result.

5. The method according to claim 4, wherein the step S1052 further comprises:

judging whether the natural language identification result is a valid result, if so, outputting the natural language identification result as a current output field; if the result is not an effective result, the natural language recognition result is used as a navigation analysis parameter to generate a navigation result; and updating the navigation analysis program according to the navigation result.

6. The method of claim 1, wherein the plurality of language fields are a first plurality of language fields; the semantic classification fields kill _ id are first type semantic classification fields kill _ id corresponding to the first type language fields.

7. The method of claim 6, further comprising: a second type of multiple language fields and a second type of semantic classification field twist _ id corresponding to the second type of language fields.

8. The method according to claim 7, wherein the step S105 further comprises:

judging whether the navigation position data is valid data or not, and if so, outputting the navigation position data; if not, the current field and the second semantic classification field kill _ id can be analyzed through corresponding semantics to obtain a second output field; and taking the second output field as a current output field.

9. An offline semantic parsing system, comprising: the semantic classification field establishing unit is configured to:

acquiring a plurality of corresponding functional semantic fields according to the plurality of language fields; establishing semantic corresponding relations between the plurality of functional semantic fields and the plurality of language fields;

in the gram grammar compiling content of the language model, creating a plurality of semantic classification fields kill _ id corresponding to the plurality of functional semantic fields; associating a plurality of semantic resolutions according to the plurality of semantic classification fields kill _ id; the plurality of semantic resolutions can obtain a determined output field;

the analysis unit includes: an analysis unit and a navigation unit; the parsing unit is configured to:

judging whether the current field has a plurality of corresponding semantic classification fields kill _ id, if so, turning to an analysis unit, and if not, turning to a navigation unit;

the analysis unit is configured to obtain a corresponding current semantic classification field kill _ id according to the current field and the semantic corresponding relation; analyzing and acquiring a current output field through corresponding semantics according to the current semantic classification field skill _ id;

the navigation unit is configured to acquire a navigation information return field according to the current field and a local navigation analysis program; the navigation analysis program can obtain corresponding navigation position data according to the current field.

10. The system of claim 9, wherein:

the parsing unit is further configured to:

acquiring a corresponding semantic classification field kill _ id according to the current field and the semantic corresponding relation;

judging whether the semantic classification field kill _ id is empty, if not, turning to the analysis unit, and if so, turning to the navigation unit;

the parsing unit is further configured to determine whether the current semantic classification field kill _ id can obtain a current output field through corresponding semantic parsing, and if so, output the current output field; if not, turning to the navigation unit;

the navigation analysis program corresponds to navigation analysis semantics; the semantic corresponding relation comprises a corresponding relation between navigation analysis semantics and language fields;

the navigation unit is also configured to obtain corresponding current function semantics according to the current field; judging whether the current function semantics is navigation analysis semantics, if so, acquiring a navigation information return field according to the current field and a local navigation analysis program; if not, identifying the current field through the natural language to obtain a current natural language identification result;

the navigation unit is also configured to judge whether the natural language identification result is a valid result, and if the natural language identification result is the valid result, the natural language identification result is output as a current output field; if the result is not an effective result, the natural language recognition result is used as a navigation analysis parameter to generate a navigation result; updating the navigation analysis program according to the navigation result;

the plurality of language fields are a first plurality of language fields; the semantic classification fields kill _ id are first type semantic classification fields kill _ id corresponding to the first type language fields;

a second type of multiple language fields and a second type of semantic classification field twist _ id corresponding to the second type of language fields;

the navigation unit is also configured to judge whether the navigation position data is valid data, and if yes, the navigation position data is output; if not, the current field and the second semantic classification field kill _ id can be analyzed through corresponding semantics to obtain a second output field; and taking the second output field as a current output field.

Technical Field

The invention relates to the field of speech recognition and application. The invention particularly relates to an offline semantic parsing method and system.

Background

Most of the existing semantic parsing schemes are based on a repeated matching scheme to determine a final semantic result. In the existing scheme, after the recognition result is obtained, a relatively correct result is obtained by sequentially matching different offline semantics.

Therefore, the existing semantic analysis scheme has the defect that on one hand, the final semantic result is a result with relatively high reliability after being established and compared in sequence, which means that even if the final semantic result is correctly identified, the final semantic result cannot be guaranteed to be the result which the user wants to obtain.

On the other hand, semantic matching and parsing need to take time, which is not obvious in time cost under the condition of less offline semantics, and as the demand of offline semantics increases, the language and skill are more and more, the parsing time also increases, the offline content is richer, the semantic parsing time is longer, and the user experience is worse.

The inventor discovers that in the process of implementing the application: the fundamental reasons for the above-mentioned defects are: the recognition result in the above technology does not contain information related to semantics, so that the semantics which are required to be obtained correctly can only be obtained by sequentially matching and comparing the offline semantics.

Disclosure of Invention

The embodiment of the invention provides an offline semantic analysis method and system, which are used for solving at least one of the technical problems.

In a first aspect, an embodiment of the present invention provides an offline semantic parsing method, including: step S101, acquiring a plurality of functional semantic fields corresponding to the plurality of language fields according to the plurality of language fields, and establishing semantic corresponding relations between the plurality of functional semantic fields and the plurality of language fields; step S102, in the gram grammar compiling content of the language model, a plurality of semantic classification fields kill _ id are created corresponding to a plurality of functional semantic fields; associating a plurality of semantic resolutions according to a plurality of semantic classification fields kill _ id; multiple semantic resolutions can obtain a determined output field; step S103, judging whether the current field has a plurality of corresponding semantic classification fields kill _ id, if so, turning to step S104, and if not, turning to step S105; step S104, acquiring a corresponding current semantic classification field kill _ id according to the corresponding relation between the current field and the semantic; analyzing and acquiring a current output field through corresponding semantics according to the current semantic classification field skill _ id; step S105, navigation information return fields are obtained according to the current fields and the local navigation analysis program; the navigation analysis program can obtain corresponding navigation position data according to the current field.

In a second aspect, an embodiment of the present invention provides an offline semantic parsing system, including: the semantic classification field establishing unit is configured to: acquiring a plurality of corresponding functional semantic fields according to the plurality of language fields; establishing semantic corresponding relations between a plurality of functional semantic fields and a plurality of language fields; in the gram grammar compiling content of the language model, a plurality of semantic classification fields kill _ id are created corresponding to a plurality of functional semantic fields; associating a plurality of semantic resolutions according to a plurality of semantic classification fields kill _ id; multiple semantic resolutions can obtain a determined output field; the analysis unit includes: a semantic analysis unit and a navigation unit. The parsing unit is configured to: judging whether the current field has a plurality of corresponding semantic classification fields kill _ id, if so, turning to an analysis unit, and if not, turning to a navigation unit; the semantic analysis unit is configured to obtain a corresponding current semantic classification field kill _ id according to the current field and the semantic corresponding relation; analyzing and acquiring a current output field through corresponding semantics according to the current semantic classification field skill _ id; the navigation unit is configured to acquire a navigation information return field according to the current field and a local navigation analysis program; the navigation analysis program can obtain corresponding navigation position data according to the current field.

In a third aspect, an electronic device for offline semantic parsing is provided, which includes: the system comprises at least one processor and a memory communicatively connected to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps of the method of any embodiment of the invention.

In a fourth aspect, the embodiments of the present invention also provide a computer program product comprising a computer program stored on a non-volatile computer-readable storage medium, the computer program comprising program instructions which, when executed by a computer, cause the computer to perform the steps of the method of any of the embodiments of the present invention.

The invention adds the field of the kill id in the flow of compiling the gram grammar, and notes the classification of each statement, thus amplifying the recognition function, and giving the information of which category the current statement belongs to while giving the recognition result: a kill id. And directly positioning the semantics of the corresponding category through the skip id for analysis.

Through the scheme, at least one defect of the following similar technologies can be overcome:

defect one: the accuracy of the semantics; because the description is marked with the class set identified by the user when the description is compiled at first, the description does not deviate to other classes for semantic analysis.

And defect two: efficiency; the sketch id is defined, so that the recognition result can be directly positioned to a corresponding semantic set, and all existing semantics cannot be sequentially traversed and compared; the problem of poor experience caused by the increase of the words and skills is not considered.

In addition, the current solution also supports and is compatible with offline navigation skills. The navigation skill as a special skill does not have the kill id as an identifier; if the recognition result has no kill id and the current product contains navigation skills, it can be determined that the current utterance is recognized by the navigation skills.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without creative efforts.

Fig. 1 is a flowchart of an offline semantic parsing method according to an embodiment of the present invention;

FIG. 2 is a flowchart of an offline semantic parsing method according to another embodiment of the present invention;

FIG. 3 is a block diagram of an offline semantic parsing method according to yet another embodiment of the present invention;

FIG. 4 is a block diagram of an offline semantic parsing system according to an embodiment of the present invention;

FIG. 5 is a flowchart of an offline semantic parsing method according to another embodiment of the present invention;

fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the embodiments of the present invention will be described clearly and completely with reference to the accompanying drawings of the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention.

The inventor finds in the process of implementing the present application that, in order to address the above-mentioned drawbacks, basic service providing capability is generally adopted, and the developers solve the drawbacks at their own discretion. Because the problem is solved, the method is accurate and quick, namely, the optimal solution is provided on the basis of not influencing the dialog creation of the developer and not increasing the extra workload of the developer. So a fusion of many underlying off-line techniques (including gram compilation, ASR, NLU, DM, etc.) is involved. Thus, there is a difficulty in the fusion of the technologies. Meanwhile, a Dialog system generally includes five components, i.e., Automatic Speech Recognition (ASR), Natural Language Understanding (NLU), Dialog Management (DM), Natural Language Generation (NLG), and Text-to-Speech synthesis (TTS). ASR is obtained through deep learning techniques and systematic improvement is difficult to make.

Referring to fig. 1, an offline semantic parsing method according to an embodiment of the present invention is shown.

As shown in fig. 1, includes:

step S101, a plurality of functional semantic fields are obtained.

In this step, a plurality of functional semantic fields corresponding to the plurality of language fields are obtained according to the plurality of language fields, and semantic corresponding relations between the plurality of functional semantic fields and the plurality of language fields are established.

Step S102, a plurality of semantic classification fields are created.

In this step, a plurality of semantic classification fields kill _ id are created corresponding to a plurality of functional semantic fields in the gram syntax compilation content of the language model. Associating a plurality of semantic resolutions according to a plurality of semantic classification fields kill _ id. Multiple semantic resolutions can obtain a determined output field.

Step S103, dividing analysis methods.

In this step, it is determined whether the current field has a plurality of semantic classification fields kill _ id, if yes, the process goes to step S104, and if not, the process goes to step S105.

And step S104, semantic parsing.

In this step, a corresponding current semantic classification field kill _ id is obtained according to the corresponding relationship between the current field and the semantic. And analyzing and acquiring the current output field according to the corresponding semantics of the current semantic classification field skill _ id.

Step S105, navigation analysis.

In this step, the navigation information return field is obtained according to the current field and the local navigation analysis program. The navigation analysis program can obtain corresponding navigation position data according to the current field.

In a preferred embodiment of the offline semantic parsing method, as shown in fig. 2, step S103 includes:

and step S1031, acquiring the current semantic classification field.

In this step, a corresponding semantic classification field kill _ id is obtained according to the current field and the semantic corresponding relationship.

Step S1032 judges whether the semantic classification field is empty.

In this step, it is determined whether the semantic classification field kill _ id is empty, and if not, the process goes to step S104, and if so, the process goes to step S105.

In another preferred embodiment of the offline semantic parsing method, step S104 includes:

and judging whether the current output field can be obtained or not through corresponding semantic analysis of the current semantic classification field kill _ id, and if so, outputting the current output field. If not, the process proceeds to step S105.

In another preferred embodiment of the offline semantic parsing method, the navigation parser corresponds to the navigation parsing semantics. The semantic corresponding relationship comprises the corresponding relationship between the navigation analysis semantic and the language field.

As shown in fig. 3, step S105 includes:

step S1051, obtaining the current functional semantic.

In this step, the corresponding current functional semantics are obtained according to the current field.

Step S1052, determining whether the current functional semantic is a navigation parsing semantic.

In the step, whether the current functional semantics is the navigation analysis semantics is judged, and if yes, the navigation information return field is obtained according to the current field and a local navigation analysis program. If not, the current natural language identification result is obtained through the natural language identification current field.

In another preferred embodiment of the offline semantic analysis method, step S1052 further includes determining whether the natural language identification result is a valid result, and if so, outputting the natural language identification result as the current output field. And if the result is not an effective result, the natural language recognition result is used as a navigation analysis parameter to generate a navigation result. And updating the navigation analysis program according to the navigation result.

In a further preferred embodiment of the offline semantic parsing method, the plurality of language fields are a first plurality of language fields. The semantic classification fields kill _ id are first type semantic classification fields kill _ id corresponding to the first type language fields.

In another preferred embodiment of the offline semantic parsing method, the method further includes: a second type of multi-language field and a second type of semantic classification field twist _ id corresponding to the second type of language field.

In another preferred embodiment of the offline semantic analysis method, step S105 further includes:

and judging whether the navigation position data is valid data or not, and if so, outputting the navigation position data. If not, the current field and the second semantic classification field skill _ id can analyze through corresponding semantics to obtain a second output field. The second output field is taken as the current output field.

In a second aspect, an embodiment of the present invention provides an offline semantic parsing system, as shown in fig. 4, including: semantic classification field establishing unit 101 and parsing unit 201, where semantic classification field establishing unit 101 is configured to:

and acquiring a plurality of corresponding functional semantic fields according to the plurality of language fields. And establishing semantic corresponding relations between the plurality of functional semantic fields and the plurality of language fields.

In the gram syntax compilation content of the language model, a plurality of semantic classification fields kill _ id are created corresponding to a plurality of functional semantic fields. Associating a plurality of semantic resolutions according to a plurality of semantic classification fields kill _ id. Multiple semantic resolutions can obtain a determined output field.

The analysis unit 201 includes: a semantic parsing unit 202 and a navigation unit 203. The parsing unit 201 is configured to:

and judging whether the current field has a plurality of corresponding semantic classification fields kill _ id, if so, turning to the analysis unit 201, and if not, turning to the navigation unit 203.

The semantic parsing unit 202 is configured to obtain a corresponding current semantic classification field kill _ id according to the current field and the semantic corresponding relationship. And analyzing and acquiring the current output field according to the corresponding semantics of the current semantic classification field skill _ id.

The navigation unit 203 is configured to obtain the navigation information return field according to the current field and the local navigation analysis program. The navigation analysis program can obtain corresponding navigation position data according to the current field.

In another preferred embodiment of the offline semantic parsing system:

the parsing unit 201 is further configured to:

and acquiring a corresponding semantic classification field kill _ id according to the current field and the semantic corresponding relation.

And judging whether the semantic classification field kill _ id is empty, if not, turning to an analysis unit 201, and if so, turning to a navigation unit 203.

The semantic parsing unit 202 is further configured to determine whether the current output field can be obtained by the corresponding semantic parsing of the current semantic classification field kill _ id, and if so, output the current output field. If not, go to

The navigation analysis program corresponds to the navigation analysis semantics. The semantic corresponding relationship comprises the corresponding relationship between the navigation analysis semantic and the language field.

The navigation unit 203 is further configured to obtain a corresponding current functional semantic according to the current field. And judging whether the current functional semantics are navigation analysis semantics, if so, acquiring a navigation information return field according to the current field and a local navigation analysis program. If not, the current natural language identification result is obtained through the natural language identification current field.

The navigation unit 203 is further configured to determine whether the natural language recognition result is a valid result, and if the natural language recognition result is a valid result, output the natural language recognition result as a current output field. And if the result is not an effective result, the natural language recognition result is used as a navigation analysis parameter to generate a navigation result. And updating the navigation analysis program according to the navigation result.

The plurality of language fields are a first plurality of language fields. The semantic classification fields kill _ id are first type semantic classification fields kill _ id corresponding to the first type language fields.

A second type of multi-language field and a second type of semantic classification field twist _ id corresponding to the second type of language field.

The navigation unit 203 is further configured to determine whether the navigation position data is valid data, and if so, output the navigation position data. If not, the current field and the second semantic classification field skill _ id can analyze through corresponding semantics to obtain a second output field. The second output field is taken as the current output field.

In other embodiments, the present invention further provides a non-volatile computer storage medium, where the computer storage medium stores computer-executable instructions, where the computer-executable instructions may perform the encryption method of the network model in any of the above method embodiments;

as one embodiment, a non-transitory computer storage medium of the present invention stores computer-executable instructions configured to:

and acquiring a plurality of corresponding functional semantic fields according to the plurality of language fields. And establishing semantic corresponding relations between the plurality of functional semantic fields and the plurality of language fields.

In the gram syntax compilation content of the language model, a plurality of semantic classification fields kill _ id are created corresponding to a plurality of functional semantic fields. Associating a plurality of semantic resolutions according to a plurality of semantic classification fields kill _ id. Multiple semantic resolutions can obtain a determined output field.

And judging whether the current field has a plurality of corresponding semantic classification fields skill _ id, if so, acquiring the corresponding current semantic classification field skill _ id according to the corresponding relation between the current field and the semantic. And analyzing and acquiring the current output field according to the corresponding semantics of the current semantic classification field skill _ id. If not, the navigation information return field is obtained according to the current field and the local navigation analysis program. The navigation analysis program can obtain corresponding navigation position data according to the current field.

In another preferred embodiment of the offline semantic parsing method, before explaining the scheme of the present invention, the following concept for the kill _ id is expanded: each term can be classified into a category, such as "turn on tv" and "turn on light", and these terms can be classified as "home control" (the name of "shop control" here is decided by the developer); for example, "how much weather today" can be classified as "weather query"; so as to be various; since each utterance can be classified under a certain set, we can add a unique identifier, which is the skip id, to the set of utterances.

The processing flow of the offline semantic analysis method in one embodiment of the present invention is shown in fig. 5, and the scheme is described as follows:

the classification of each statement is noted by adding a kill id field in the flow of gram grammar compiling, so that the recognition function is amplified, and information of which category the current statement belongs to is given while a recognition result is given: a kill id. And directly positioning the semantics of the corresponding category through the skip id for analysis.

Pcm represents outputting the currently recognized speech, as shown in fig. 5. And acquiring an identification result asr feed. And judging the corresponding identification type asr _ ret.Skillid as nil according to the identification result asr feed, namely judging whether the identification type is null, and if not, acquiring a result field or result data corresponding to the identification type, namely, dui _ nul feed askillid. And after obtaining the result field or the result data, judging whether the result field or the result data is a result feed result which can be output, if so, outputting the result and ending the identification process. If not, selecting to execute the navigation skill.

If the recognition category asr _ ret.skillid is nil, that is, the recognition category is null, it indicates that the current recognized speech is not in the preset recognition category. If the category is empty, then the "navigation skill" is selected to be performed. Judging whether the current recognition voice is in the 'navigation skill' currkill is navi _ kill, if so, using navi _ nlu feed as the 'navigation skill' to return a navigation result; if not, the recognition mode of the natural language is used for recognizing the current recognition voice, and the recognition result is output.

And further, judging whether the navigation skill returned navigation result can be obtained, if so, outputting, otherwise, adopting other recognition categories to recognize the recognition voice, acquiring an identification result is fed result ok, and returning an output result return.

Through the scheme, the following defects of similar technologies can be obviously solved:

defect one: the accuracy of the semantics; because the description is marked with the class set identified by the user when the description is compiled at first, the description does not deviate to other classes for semantic analysis.

And defect two: efficiency; the sketch id is defined, so that the recognition result can be directly positioned to a corresponding semantic set, and all existing semantics cannot be sequentially traversed and compared; the problem of poor experience caused by the increase of the words and skills is not considered.

In addition, the current solution also supports and is compatible with offline navigation skills. The navigation skill as a special skill does not have the kill id as an identifier; if the recognition result has no kill id and the current product contains navigation skills, it can be determined that the current expression is the navigation skills.

Example one:

the user: turning on a television

ASR: and (3) recognition results: turning on a television; stick _ id: control of household appliances

NLU: locating to home appliance controls for semantic parsing

DUI: good, this opens for you.

Example two:

the user: navigating to Suzhou

ASR: and (3) recognition results: turning on a television; stick _ id: ' (No mark, identified as navigation skill)

NLU: positioning navigation skill

DUI: well, this is just your navigation.

As a non-volatile computer-readable storage medium, the non-volatile software program, the non-volatile computer-executable program, and modules, such as program instructions/modules corresponding to the offline semantic parsing method according to the embodiments of the present invention, may be stored. One or more program instructions are stored in a non-transitory computer readable storage medium that, when executed by a processor, perform the offline semantic parsing method of any of the method embodiments described above.

The non-volatile computer-readable storage medium may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to use of an offline semantic parsing apparatus, and the like. Further, the non-volatile computer-readable storage medium may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, the non-transitory computer readable storage medium optionally includes memory located remotely from the processor, which may be connected to the offline semantic parsing device through a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.

Embodiments of the present invention also provide a computer program product, which includes a computer program stored on a non-volatile computer-readable storage medium, where the computer program includes program instructions, which, when executed by a computer, cause the computer to perform any one of the above-mentioned offline semantic parsing methods.

Fig. 6 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, and as shown in fig. 6, the electronic device includes: one or more processors 610 and a memory 620, with one processor 610 being an example in fig. 6. The device of the offline semantic parsing method may further include: an input device 630 and an output device 640. The processor 610, the memory 620, the input device 630, and the output device 640 may be connected by a bus or other means, such as the bus connection in fig. 6. The memory 620 is a non-volatile computer-readable storage medium as described above. The processor 610 executes various functional applications of the server and data processing by running nonvolatile software programs, instructions and modules stored in the memory 620, that is, the offline semantic parsing method of the above method embodiment is implemented. The input device 630 may receive input numeric or character information and generate key signal inputs related to user settings and function controls of the information delivery device. The output device 640 may include a display device such as a display screen.

The product can execute the method provided by the embodiment of the invention, and has corresponding functional modules and beneficial effects of the execution method. For technical details that are not described in detail in this embodiment, reference may be made to the method provided by the embodiment of the present invention.

As an embodiment, the electronic device may be applied to an encryption and decryption platform, and includes: at least one processor; and a memory communicatively coupled to the at least one processor; wherein the memory stores instructions executable by the at least one processor to enable the at least one cryptographic processor to:

and acquiring a plurality of corresponding functional semantic fields according to the plurality of language fields. Establishing semantic corresponding relations between a plurality of functional semantic fields and a plurality of language fields;

in the gram grammar compiling content of the language model, a plurality of semantic classification fields kill _ id are created corresponding to a plurality of functional semantic fields; associating a plurality of semantic resolutions according to a plurality of semantic classification fields kill _ id; multiple semantic resolutions can obtain a determined output field;

judging whether the current field has a plurality of corresponding semantic classification fields skill _ id, if so, acquiring the corresponding current semantic classification field skill _ id according to the corresponding relation between the current field and the semantic; analyzing and acquiring a current output field through corresponding semantics according to the current semantic classification field skill _ id; if not, acquiring a navigation information return field according to the current field and a local navigation analysis program; the navigation analysis program can obtain corresponding navigation position data according to the current field.

The electronic device of the embodiments of the present application exists in various forms, including but not limited to:

(1) a mobile communication device: such devices are characterized by mobile communications capabilities and are primarily targeted at providing voice, data communications. Such terminals include smart phones, multimedia phones, functional phones, and low-end phones, among others.

(2) Ultra mobile personal computer device: the equipment belongs to the category of personal computers, has calculation and processing functions and generally has the characteristic of mobile internet access. Such terminals include: PDA, MID, and UMPC devices, etc.

(3) A portable entertainment device: such devices can display and play multimedia content. The devices comprise audio and video players, handheld game consoles, electronic books, intelligent toys and portable vehicle-mounted navigation devices.

(4) The server is similar to a general computer architecture, but has higher requirements on processing capability, stability, reliability, safety, expandability, manageability and the like because of the need of providing highly reliable services.

(5) And other electronic devices with data interaction functions.

The above-described embodiments of the apparatus are merely schematic, where units illustrated as separate components may or may not be physically separate, and components shown as units may or may not be physical units, may be located in one place, or may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the embodiment. One of ordinary skill in the art can understand and implement it without inventive effort.

Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a necessary general hardware platform, and certainly can also be implemented by hardware. With this understanding, the above technical solutions may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes several instructions for enabling a computer device (which may be a personal computer, a server, or a network device, etc.) to execute the methods of the embodiments or some parts of the embodiments.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solutions of the present invention, and not for limiting the same; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that: the technical solutions described in the foregoing embodiments may be modified or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.