CN110189751A - Method of speech processing and equipment - Google Patents
Method of speech processing and equipment Download PDFInfo
- Publication number
- CN110189751A CN110189751A CN201910335969.7A CN201910335969A CN110189751A CN 110189751 A CN110189751 A CN 110189751A CN 201910335969 A CN201910335969 A CN 201910335969A CN 110189751 A CN110189751 A CN 110189751A
- Authority
- CN
- China
- Prior art keywords
- art text
- words art
- text
- word segmentation
- sentence element
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/06—Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice
- G10L15/063—Training
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G10L15/14—Speech classification or search using statistical models, e.g. Hidden Markov Models [HMMs]
- G10L15/142—Hidden Markov Models [HMMs]
- G10L15/144—Training of HMMs
-
- G—PHYSICS
- G10—MUSICAL INSTRUMENTS; ACOUSTICS
- G10L—SPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
- G10L15/00—Speech recognition
- G10L15/08—Speech classification or search
- G10L15/16—Speech classification or search using artificial neural networks
Abstract
The embodiment of the present invention provides a kind of method of speech processing and equipment, this method comprises: determining that the corresponding target of voice to be processed talks about art text；Word segmentation processing is carried out to target words art text；According to neural network model to after word segmentation processing art text carry out sentence element analysis, obtain word segmentation processing after if the corresponding sentence element of art text, the neural network model according to words art text and sentence element training obtain；According to the sentence element of acquisition, the corresponding effective words art text of the voice to be processed is determined.Method provided in this embodiment can filter out meaningless information, carry out intention assessment based on effectively words art text, reduce user's intention assessment difficulty, improve the accuracy rate of intention assessment result, while time saving and energy saving, be suitble to practical application.
Description
Technical field
The present embodiments relate to field of computer technology more particularly to a kind of method of speech processing and equipment.
Background technique
With economic continuous development, the communication technology is developed rapidly, and more and more people begin to use communication to transport
The communication system of battalion quotient communicates.
Currently, user when using the communication system of communication operator, if what problem encountered, usually transports to communication
Seek the customer service system consulting of quotient.Customer service system needs to identify that existing customer service system is generally directly based upon use to user's intention
The inquiry at family carries out user's intention assessment.
However, some meaningless information, such as colloquial expression are inevitably had in the inquiry of user, such as:
" uh ", " Kazakhstan ", modal particles such as " " virtually can increase difficulty for user's intention assessment, reduce the standard of intention assessment result
True rate.
Summary of the invention
The embodiment of the present invention provides a kind of method of speech processing and equipment, to overcome existing customer service system to be directly based upon user
Inquiry carry out user's intention assessment, identification difficulty is big, accuracy rate is low problem.
In a first aspect, the embodiment of the present invention provides a kind of method of speech processing, comprising:
Determine the corresponding target words art text of voice to be processed；
Word segmentation processing is carried out to target words art text；
Sentence element analysis is carried out to art text after word segmentation processing according to neural network model, after obtaining word segmentation processing
If the corresponding sentence element of art text, the neural network model according to words art text and sentence element training obtain；
According to the sentence element of acquisition, the corresponding effective words art text of the voice to be processed is determined.
In a kind of possible design, the sentence element according to acquisition, determining that the voice to be processed is corresponding has
Effect words art text, comprising:
Subject, predicate, object, attribute, the adverbial modifier and complement in the sentence element of acquisition is identified, is tied according to identification
Fruit is filtered the sentence element of acquisition；
The corresponding effective words art text of the voice to be processed is obtained according to filter result.
In a kind of possible design, above-mentioned method, further includes:
Intention assessment is carried out according to effective words art text.
It is described that intention assessment is carried out according to effective words art text in a kind of possible design, comprising:
Intention keyword extraction is carried out to effective words art text according to default intention keyword；
Intention assessment is carried out according to the intention keyword of extraction.
In a kind of possible design, the corresponding target of the determination voice to be processed talks about art text, comprising:
Speech recognition is carried out to the voice to be processed, obtains the target words art text；
The method also includes:
Remove the preset characters in target words art text, and will remove the target words art text after preset characters as
New target talks about art text, executes described the step of carrying out word segmentation processing to target words art text.
It is described that word segmentation processing is carried out to target words art text in a kind of possible design, comprising:
Participle model based on condition random field or hidden Markov carries out word segmentation processing to target words art；
The method also includes:
Art text if judging whether there is after word segmentation processing；
Art text after word segmentation processing if it exists, then execute it is described according to neural network model to after word segmentation processing
Art text carries out the step of sentence element analysis；
Art text after word segmentation processing if it does not exist then generates participle and unsuccessfully prompts.
Second aspect, the embodiment of the present invention provide a kind of speech processing device, including memory, processor and are stored in
In the memory and the computer executed instructions that can run on the processor, the processor execute the computer and hold
Following steps are realized when row instruction:
Determine the corresponding target words art text of voice to be processed；
Word segmentation processing is carried out to target words art text；
Sentence element analysis is carried out to art text after word segmentation processing according to neural network model, after obtaining word segmentation processing
If the corresponding sentence element of art text, the neural network model according to words art text and sentence element training obtain；
According to the sentence element of acquisition, the corresponding effective words art text of the voice to be processed is determined.
In a kind of possible design, the sentence element according to acquisition, determining that the voice to be processed is corresponding has
Effect words art text, comprising:
Subject, predicate, object, attribute, the adverbial modifier and complement in the sentence element of acquisition is identified, is tied according to identification
Fruit is filtered the sentence element of acquisition；
The corresponding effective words art text of the voice to be processed is obtained according to filter result.
In a kind of possible design, the processor realizes following steps when executing the computer executed instructions:
Intention assessment is carried out according to effective words art text.
It is described that intention assessment is carried out according to effective words art text in a kind of possible design, comprising:
Intention keyword extraction is carried out to effective words art text according to default intention keyword；
Intention assessment is carried out according to the intention keyword of extraction.
In a kind of possible design, the corresponding target of the determination voice to be processed talks about art text, comprising:
Speech recognition is carried out to the voice to be processed, obtains the target words art text；
The processor realizes following steps when executing the computer executed instructions:
Remove the preset characters in target words art text, and will remove the target words art text after preset characters as
New target talks about art text, executes described the step of carrying out word segmentation processing to target words art text.
It is described that word segmentation processing is carried out to target words art text in a kind of possible design, comprising:
Participle model based on condition random field or hidden Markov carries out word segmentation processing to target words art；
The processor realizes following steps when executing the computer executed instructions:
Art text if judging whether there is after word segmentation processing；
Art text after word segmentation processing if it exists, then execute it is described according to neural network model to after word segmentation processing
Art text carries out the step of sentence element analysis；
Art text after word segmentation processing if it does not exist then generates participle and unsuccessfully prompts.
The third aspect, the embodiment of the present invention provide a kind of computer readable storage medium, the computer-readable storage medium
Computer executed instructions are stored in matter, when processor execute the computer executed instructions when, realize first aspect as above with
And method of speech processing described in the various possible designs of first aspect.
Method of speech processing provided in this embodiment and equipment, this method is by determining the corresponding target words of voice to be processed
Art text carries out word segmentation processing to target words art text, is carried out according to neural network model to art text after word segmentation processing
Sentence element is analyzed, the corresponding sentence element of art text after acquisition word segmentation processing, wherein neural network model is according to words art
Text and sentence element training obtain, and finally further according to the sentence element of acquisition, determine the corresponding effective words art of voice to be processed
Text filters out meaningless information, carries out intention assessment based on effectively words art text, reduces user's intention assessment difficulty, mention
The accuracy rate of high intention assessment result, at the same it is time saving and energy saving, it is suitble to practical application.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is this hair
Bright some embodiments for those of ordinary skill in the art without any creative labor, can be with
It obtains other drawings based on these drawings.
Fig. 1 is the application scenario diagram of method of speech processing provided in an embodiment of the present invention；
Fig. 2 is the flow diagram one of method of speech processing provided in an embodiment of the present invention；
Fig. 3 is the flow diagram two of method of speech processing provided in an embodiment of the present invention；
Fig. 4 is the structural schematic diagram one of speech processing device provided in an embodiment of the present invention；
Fig. 5 is the structural schematic diagram two of speech processing device provided in an embodiment of the present invention；
Fig. 6 is the hardware structural diagram of speech processing device provided in an embodiment of the present invention.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
Description and claims of this specification and term " first ", " second ", " third " " in above-mentioned attached drawing
The (if present)s such as four " are to be used to distinguish similar objects, without being used to describe a particular order or precedence order.It should manage
The data that solution uses in this way are interchangeable under appropriate circumstances, so that the embodiment of the present invention described herein for example can be to remove
Sequence other than those of illustrating or describe herein is implemented.In addition, term " includes " and " having " and theirs is any
Deformation, it is intended that cover it is non-exclusive include, for example, containing the process, method of a series of steps or units, system, production
Product or equipment those of are not necessarily limited to be clearly listed step or unit, but may include be not clearly listed or for this
A little process, methods, the other step or units of product or equipment inherently.
With economic continuous development, the communication technology is developed rapidly, and more and more people begin to use communication to transport
The communication system of battalion quotient communicates.Currently, user is asked when using the communication system of communication operator if what is encountered
Topic, usually to the customer service system consulting of communication operator.Customer service system need to user be intended to identify, existing customer service system
The inquiry that user is directly based upon as unified carries out user's intention assessment.However, inevitably being had in the inquiry of user
Meaningless information, such as colloquial expression, such as: " uh ", " Kazakhstan ", modal particles such as " " virtually can be intended to know for user
Not Zeng Jia difficulty, reduce the accuracy rate of intention assessment result.
Accordingly, it is considered to arrive the above problem, the present invention provides a kind of method of speech processing, corresponding by determination voice to be processed
Target talk about art text, to target words art text carry out word segmentation processing, according to neural network model to art after word segmentation processing
Text carry out sentence element analysis, obtain word segmentation processing after if the corresponding sentence element of art text, wherein neural network model
It is obtained according to words art text and sentence element training, finally further according to the sentence element of acquisition, determines that voice to be processed is corresponding
Effectively words art text, filters out meaningless information, carries out intention assessment based on effectively words art text, reduces user's intention assessment
Difficulty improves the accuracy rate of intention assessment result, while time saving and energy saving, is suitble to practical application.
Fig. 1 is a kind of application scenario diagram of method of speech processing provided by the invention.As shown in Figure 1, customer service system 101 can
Art text is talked about with the corresponding target of determination voice to be processed, art text can be talked about to target and carry out word segmentation processing, it can also basis
Neural network model 102 to after word segmentation processing art text carry out sentence element analysis, obtain word segmentation processing after if art text
This corresponding sentence element, wherein neural network model 102 is obtained according to words art text and sentence element training, can be with root
According to the sentence element of acquisition, the corresponding effective words art text of the voice to be processed is determined.
Wherein, customer service system can provide dialogue platform for user, engage in the dialogue with user, there is processing business to seek advice from,
The functions such as complaint.
Fig. 2 is the flow diagram one of method of speech processing provided in an embodiment of the present invention, the executing subject of the present embodiment
It can be the customer service system in embodiment illustrated in fig. 1, or other equipment, such as terminal, processor, server etc., this
Embodiment is not particularly limited herein.As shown in Fig. 2, this method may include:
S201, the corresponding target words art text of voice to be processed is determined.
Wherein, voice to be processed can be determines according to actual conditions one or more voices to be treated.To
The voice of business consultation, complaint etc. can be carried out for user by handling voice.
Optionally, the corresponding target of determination voice to be processed talks about art text, comprising:
Speech recognition is carried out to the voice to be processed, obtains the target words art text；
The method also includes:
Remove the preset characters in target words art text, and will remove the target words art text after preset characters as
New target talks about art text, executes described the step of carrying out word segmentation processing to target words art text.
Specifically, speech recognition technology, also referred to as automatic speech recognition Automatic Speech Recognition,
Abbreviation ASR, target are that the vocabulary Content Transformation in the voice by the mankind is computer-readable input.
Speech recognition is carried out to above-mentioned voice to be processed, corresponding target words art text is obtained, can also further remove
Preset characters in the target words art text, wherein preset characters can be set according to actual needs, such as removal predetermined word
Mother, number etc. remove meaningless information, improve the accuracy rate of subsequent intention assessment.
In addition, other can also be arranged according to the actual situation in addition to the preset characters in above-mentioned removal target words art text
Pretreatment mode, such as normalized, normalization refer to a series of relevant tasks, can be placed on all texts same
On horizontal zone: all texts being converted to same example, number is converted into corresponding text etc..
S202, word segmentation processing is carried out to target words art text.
Here, after the corresponding target words art text of above-mentioned determination voice to be processed, target words art text can also be stored
This, can also talk about art text with displaying target, facilitate related personnel to check, audit corresponding informance.
It is optionally, described that word segmentation processing is carried out to target words art text, comprising:
Participle model based on condition random field or hidden Markov carries out word segmentation processing to target words art；
The method also includes:
Art text if judging whether there is after word segmentation processing；
Art text after word segmentation processing if it exists, then execute it is described according to neural network model to after word segmentation processing
Art text carries out the step of sentence element analysis；
Art text after word segmentation processing if it does not exist then generates participle and unsuccessfully prompts.
Specifically, condition random field (Conditional Random Field, abbreviation CRF), is a kind of machine learning skill
Art participle, CRF assign participle as the lexeme classification problem of word, and the lexeme information of usual defined word is as follows: prefix, commonly using B indicates；
In word, commonly using M is indicated；Suffix, commonly using E indicates；List word, commonly use S indicate, CRF participle process be exactly to lexeme mark after,
By the word and S individual character composition participle between B and E, such as original example sentence: after I likes that Beijing Tian An-men, CRF mark: I/S
Love/the north the S/capital B/E days/B peace/M/E, word segmentation result: I/love/Beijing/Tian An-men.
Hidden Markov model (Hidden Markov Model, abbreviation HMM) is statistical model, it is used to describe one
Markov process containing implicit unknown parameter.Its state cannot observe directly, but can be seen by observation vector sequence
It observes, each observation vector is to show as various states by certain probability density distributions, each observation vector is by one
A status switch with corresponding probability density distribution generates.So hidden Markov model is a dual random mistake
Journey ----Hidden Markov Chain and display random function collection with certain status number.
In addition, except the above-mentioned participle model based on condition random field or hidden Markov carries out target words art text
Outside word segmentation processing, other participle modes, such as word-based n-gram model can also be used according to the actual situation, to described
Target talks about art text and carries out word segmentation processing, meets plurality of application scenes needs.
In embodiments of the present invention, after carrying out word segmentation processing to target words art text, participle is judged whether there is
Treated otherwise words art text, generates participle and unsuccessfully prompts, related personnel is facilitated to look into if it is present executing subsequent step
It sees information and carries out respective handling.
S203, sentence element analysis is carried out to art text after word segmentation processing according to neural network model, is segmented
Treated the corresponding sentence element of words art text, the neural network model are trained according to words art text and sentence element
It arrives.
Here, neural network (Neural Networks, abbreviation NN) is (referred to as refreshing by a large amount of, simple processing unit
Through member) widely interconnect and the complex networks system that is formed, it reflects many essential characteristics of human brain function, is one
Highly complex non-linear dynamic learning system.Neural network have large-scale parallel, distributed storage and processing, self-organizing,
Adaptive and self-learning ability is particularly suitable for processing and needs while considering many factors and condition, inaccurate and fuzzy information
Processing problem.
In embodiments of the present invention, above-mentioned neural network model is obtained according to words art text and sentence element training, according to
Trained neural network model carries out sentence element analysis to art text after word segmentation processing, obtains after word segmentation processing
The corresponding sentence element of art text.
Wherein, the constituent of sentence is also syntactic constituent sentence element.Have in sentence, between word and word certain
Syntagmatic sentence can be divided into different constituents according to different relationships.Sentence element is filled by word or phrase
When, such as the general sentence element of Chinese has: subject, predicate, object, dynamic language, attribute, the adverbial modifier, complement and head etc..
S204, the sentence element according to acquisition determine the corresponding effective words art text of the voice to be processed.
Optionally, the sentence element according to acquisition determines the corresponding effective words art text of the voice to be processed, packet
It includes:
Subject, predicate, object, attribute, the adverbial modifier and complement in the sentence element of acquisition is identified, is tied according to identification
Fruit is filtered the sentence element of acquisition；
The corresponding effective words art text of the voice to be processed is obtained according to filter result.
Here it is possible to subject, predicate, object, attribute, the adverbial modifier, the complement in sentence are identified, non-above-mentioned six kinds
Sentence structure is defaulted as meaningless vocabulary and filters this out, and the vocabulary left can form newly term sentence, will clean
Sentence afterwards is brought into subsequent intention assessment process and carries out intention assessment, reduces subsequent intention assessment difficulty.
Optionally, above-mentioned method, further includes:
Intention assessment is carried out according to effective words art text.
Method of speech processing provided in this embodiment talks about art text by the corresponding target of determination voice to be processed, to mesh
Mark words art text carries out word segmentation processing, carries out sentence element point to art text after word segmentation processing according to neural network model
It analyses, the corresponding sentence element of art text after acquisition word segmentation processing, wherein neural network model is according to words art text and sentence
Ingredient training obtains, and finally further according to the sentence element of acquisition, determines the corresponding effective words art text of voice to be processed, filters out
Meaningless information carries out intention assessment based on effectively words art text, reduces user's intention assessment difficulty, improve intention assessment knot
The accuracy rate of fruit, at the same it is time saving and energy saving, it is suitble to practical application.
Fig. 3 is the flow diagram two of method of speech processing provided in an embodiment of the present invention, and the present embodiment is in Fig. 2 embodiment
On the basis of, the specific implementation process of the present embodiment is described in detail.As shown in figure 3, this method comprises:
S301, speech recognition is carried out to voice to be processed, obtains the target words art text.
Preset characters in S302, the removal target words art text, talk about art text for the target after removal preset characters
Art text is talked about as new target, executes the step of word segmentation processing is carried out to target words art text.
S303, the participle model based on condition random field or hidden Markov carry out word segmentation processing to target words art.
S304, judge whether there is after word segmentation processing if art text.
Art text after word segmentation processing if it exists, thens follow the steps S305 to S309, if it does not exist after word segmentation processing
Art text is talked about, S310 is thened follow the steps.
S305, if it exists art text after word segmentation processing, then according to neural network model to art after word segmentation processing
Text carry out sentence element analysis, obtain word segmentation processing after if the corresponding sentence element of art text, the neural network model
It is obtained according to words art text and sentence element training.
S306, the subject in the sentence element of acquisition, predicate, object, attribute, the adverbial modifier and complement identify, according to
Recognition result is filtered the sentence element of acquisition.
S307, the corresponding effective words art text of the voice to be processed is obtained according to filter result.
S308, intention keyword extraction is carried out to effective words art text according to default intention keyword.
S309, intention assessment is carried out according to the intention keyword of extraction.
Here, presetting intention keyword is the keyword that can be indicated user and be intended to, and can be arranged according to the actual situation.It is logical
Default intention keyword is crossed to above-mentioned effective words art text progress intention keyword extraction, further according to extraction intention keyword into
Row intention assessment, simply, conveniently, and intention assessment accuracy rate is high, is suitble to application.
S310, if it does not exist art text after word segmentation processing, then generate participle and unsuccessfully prompt.
Method of speech processing provided in this embodiment, to user, term sentence carries out sentence before entering intention assessment process
Subconstiuent analysis, filters out meaningless information, carries out intention assessment based on effectively words art text, it is difficult to reduce user's intention assessment
Degree improves the accuracy rate of intention assessment result, while time saving and energy saving, is suitble to practical application.
Fig. 4 is the structural schematic diagram one of speech processing device provided in an embodiment of the present invention.As shown in figure 4, at the voice
Reason equipment 40 includes: that text determining module 401, text word segmentation module 402, Sentence analysis module 403 and effectively words art determine
Module 404.
Text determining module 401, for determining the corresponding target words art text of voice to be processed.
Text word segmentation module 402, for carrying out word segmentation processing to target words art text.
Sentence analysis module 403, for according to neural network model to after word segmentation processing art text carry out sentence at
Point analysis, obtain word segmentation processing after if the corresponding sentence element of art text, the neural network model according to words art text and
Sentence element training obtains.
Effectively words art determining module 404, for the sentence element according to acquisition, determining that the voice to be processed is corresponding has
Effect words art text.
Equipment provided in this embodiment can be used for executing the technical solution of above method embodiment, realization principle and skill
Art effect is similar, and details are not described herein again for the present embodiment.
Fig. 5 is the structural schematic diagram two of speech processing device provided in an embodiment of the present invention.As shown in figure 5, the present embodiment
On the basis of Fig. 4 embodiment, further includes: intention assessment module 405, text judgment module 406 and participle cue module 407.
In a kind of possible design, effective words art determining module 404 includes 4041 He of sentence element recognition unit
Effectively words art obtaining unit 4042.
Wherein, sentence element recognition unit 4041, for the subject, predicate, object, fixed in the sentence element to acquisition
Language, the adverbial modifier and complement identify, are filtered according to sentence element of the recognition result to acquisition.
Effectively words art obtaining unit 4042, for obtaining the corresponding effective words art of the voice to be processed according to filter result
Text.
In a kind of possible design, it is intended that identification module 405, for carrying out intention knowledge according to effective words art text
Not.
In a kind of possible design, the intention assessment module 405 includes keyword extraction unit 4051 and is intended to know
Other unit 4052.
Wherein, keyword extraction unit 4051, for being carried out according to default intention keyword to effective words art text
Intention keyword extracts.
Intention assessment unit 4052, for carrying out intention assessment according to the intention keyword of extraction.
In a kind of possible design, the text determining module 401 is also used to carry out voice to the voice to be processed
Identification obtains the target words art text.
The text word segmentation module 402 is also used to remove the preset characters in the target words art text, and will remove pre-
If the target words art text after character talks about art text as new target, execution is described to segment target words art text
The step of processing.
In a kind of possible design, the text word segmentation module 402 is also used to based on condition random field or hidden Ma Erke
The participle model of husband carries out word segmentation processing to target words art.
Text judgment module 406, the art text for judging whether there is after word segmentation processing.
The Sentence analysis module 403, art text if being also used to after word segmentation processing if it exists, then execute described according to mind
Art text carries out the step of sentence element analysis after network model is to word segmentation processing.
Cue module 407 is segmented, for art text after word segmentation processing if it does not exist, then participle is generated and unsuccessfully prompts.
Equipment provided in this embodiment can be used for executing the technical solution of above method embodiment, realization principle and skill
Art effect is similar, and details are not described herein again for the present embodiment.
Fig. 6 is the hardware structural diagram of speech processing device provided in an embodiment of the present invention.As shown in fig. 6, this implementation
The speech processing device 60 of example includes: processor 601 and memory 602；Wherein
Memory 602, for storing computer executed instructions；
Processor 601, for executing the computer executed instructions of memory storage, to realize in above-described embodiment at voice
Each step performed by reason method.It specifically may refer to the associated description in preceding method embodiment.
Optionally, memory 602 can also be integrated with processor 601 either independent.
When memory 602 is independently arranged, which further includes bus 603, for connecting the memory
602 and processor 601.
The embodiment of the present invention also provides a kind of computer readable storage medium, stores in the computer readable storage medium
There are computer executed instructions, when processor executes the computer executed instructions, realizes method of speech processing as described above.
In several embodiments provided by the present invention, it should be understood that disclosed device and method can pass through it
Its mode is realized.For example, apparatus embodiments described above are merely indicative, for example, the division of the module, only
Only a kind of logical function partition, there may be another division manner in actual implementation, for example, multiple modules can combine or
It is desirably integrated into another system, or some features can be ignored or not executed.Another point, it is shown or discussed it is mutual it
Between coupling, direct-coupling or communication connection can be through some interfaces, the INDIRECT COUPLING or communication link of device or module
It connects, can be electrical property, mechanical or other forms.
The module as illustrated by the separation member may or may not be physically separated, aobvious as module
The component shown may or may not be physical unit, it can and it is in one place, or may be distributed over multiple
In network unit.Some or all of the modules therein can be selected to realize the mesh of this embodiment scheme according to the actual needs
's.
It, can also be in addition, each functional module in each embodiment of the present invention can integrate in one processing unit
It is that modules physically exist alone, can also be integrated in one unit with two or more modules.Above-mentioned module at
Unit both can take the form of hardware realization, can also realize in the form of hardware adds SFU software functional unit.
The above-mentioned integrated module realized in the form of software function module, can store and computer-readable deposit at one
In storage media.Above-mentioned software function module is stored in a storage medium, including some instructions are used so that a computer
Equipment (can be personal computer, server or the network equipment etc.) or processor (English: processor) execute this Shen
Please each embodiment the method part steps.
It should be understood that above-mentioned processor can be central processing unit (Central Processing Unit, abbreviation CPU),
It can also be other general processors, digital signal processor (Digital Signal Processor, abbreviation DSP), dedicated
Integrated circuit (Application Specific Integrated Circuit, abbreviation ASIC) etc..General processor can be
Microprocessor or the processor are also possible to any conventional processor etc..It can be in conjunction with the step of invention disclosed method
Be embodied directly in hardware processor and execute completion, or in processor hardware and software module combination execute completion.
Memory may include high speed RAM memory, it is also possible to and it further include non-volatile memories NVM, for example, at least one
Magnetic disk storage can also be USB flash disk, mobile hard disk, read-only memory, disk or CD etc..
It is total that bus can be industry standard architecture (Industry Standard Architecture, abbreviation ISA)
Line, external equipment interconnection (Peripheral Component, abbreviation PCI) bus or extended industry-standard architecture
(Extended Industry Standard Architecture, abbreviation EISA) bus etc..It is total that bus can be divided into address
Line, data/address bus, control bus etc..For convenient for indicating, the bus in illustrations does not limit an only bus or one
The bus of seed type.
Above-mentioned storage medium can be by any kind of volatibility or non-volatile memory device or their combination
It realizes, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable
Read-only memory (EPROM), programmable read only memory (PROM), read-only memory (ROM), magnetic memory, flash memory,
Disk or CD.Storage medium can be any usable medium that general or specialized computer can access.
A kind of illustrative storage medium is coupled to processor, believes to enable a processor to read from the storage medium
Breath, and information can be written to the storage medium.Certainly, storage medium is also possible to the component part of processor.It processor and deposits
Storage media can be located at specific integrated circuit (Application Specific Integrated Circuits, abbreviation ASIC)
In.Certainly, pocessor and storage media can also be used as discrete assembly and be present in electronic equipment or main control device.
Those of ordinary skill in the art will appreciate that: realize that all or part of the steps of above-mentioned each method embodiment can lead to
The relevant hardware of program instruction is crossed to complete.Program above-mentioned can be stored in a computer readable storage medium.The journey
When being executed, execution includes the steps that above-mentioned each method embodiment to sequence；And storage medium above-mentioned include: ROM, RAM, magnetic disk or
The various media that can store program code such as person's CD.
Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention., rather than its limitations；To the greatest extent
Pipe present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that: its according to
So be possible to modify the technical solutions described in the foregoing embodiments, or to some or all of the technical features into
Row equivalent replacement；And these are modified or replaceed, various embodiments of the present invention technology that it does not separate the essence of the corresponding technical solution
The range of scheme.
Claims (11)
1. a kind of method of speech processing characterized by comprising
Determine the corresponding target words art text of voice to be processed；
Word segmentation processing is carried out to target words art text；
Sentence element analysis is carried out to art text after word segmentation processing according to neural network model, is obtained after word segmentation processing
The corresponding sentence element of art text, the neural network model are obtained according to words art text and sentence element training；
According to the sentence element of acquisition, the corresponding effective words art text of the voice to be processed is determined.
2. the method according to claim 1, wherein the sentence element according to acquisition, determines described wait locate
Manage the corresponding effective words art text of voice, comprising:
Subject, predicate, object, attribute, the adverbial modifier and complement in the sentence element of acquisition is identified, according to recognition result pair
The sentence element of acquisition is filtered；
The corresponding effective words art text of the voice to be processed is obtained according to filter result.
3. method according to claim 1 or 2, which is characterized in that further include:
Intention assessment is carried out according to effective words art text.
4. according to the method described in claim 3, it is characterized in that, described carry out intention knowledge according to effective words art text
Not, comprising:
Intention keyword extraction is carried out to effective words art text according to default intention keyword；
Intention assessment is carried out according to the intention keyword of extraction.
5. the method according to claim 1, wherein the corresponding target words art text of determination voice to be processed
This, comprising:
Speech recognition is carried out to the voice to be processed, obtains the target words art text；
The method also includes:
It removes the preset characters in target words art text, and target words art text after preset characters will be removed as newly
Target talks about art text, executes described the step of carrying out word segmentation processing to target words art text.
6. the method according to claim 1, wherein it is described to the target words art text carry out word segmentation processing,
Include:
Participle model based on condition random field or hidden Markov carries out word segmentation processing to target words art；
The method also includes:
Art text if judging whether there is after word segmentation processing；
Art text after word segmentation processing if it exists then executes described literary to art after word segmentation processing according to neural network model
The step of this progress sentence element analysis；
Art text after word segmentation processing if it does not exist then generates participle and unsuccessfully prompts.
7. a kind of speech processing device, which is characterized in that in the memory and can including memory, processor and storage
The computer executed instructions run on the processor, the processor are realized as follows when executing the computer executed instructions
Step:
Determine the corresponding target words art text of voice to be processed；
Word segmentation processing is carried out to target words art text；
Sentence element analysis is carried out to art text after word segmentation processing according to neural network model, is obtained after word segmentation processing
The corresponding sentence element of art text, the neural network model are obtained according to words art text and sentence element training；
According to the sentence element of acquisition, the corresponding effective words art text of the voice to be processed is determined.
8. equipment according to claim 7, which is characterized in that the sentence element according to acquisition determines described wait locate
Manage the corresponding effective words art text of voice, comprising:
Subject, predicate, object, attribute, the adverbial modifier and complement in the sentence element of acquisition is identified, according to recognition result pair
The sentence element of acquisition is filtered；
The corresponding effective words art text of the voice to be processed is obtained according to filter result.
9. equipment according to claim 7 or 8, which is characterized in that the processor executes the computer executed instructions
Shi Shixian following steps:
Intention assessment is carried out according to effective words art text.
10. equipment according to claim 9, which is characterized in that described to carry out intention knowledge according to effective words art text
Not, comprising:
Intention keyword extraction is carried out to effective words art text according to default intention keyword；
Intention assessment is carried out according to the intention keyword of extraction.
11. a kind of computer readable storage medium, which is characterized in that be stored with computer in the computer readable storage medium
It executes instruction, when processor executes the computer executed instructions, realizes such as voice as claimed in any one of claims 1 to 6
Processing method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910335969.7A CN110189751A (en) | 2019-04-24 | 2019-04-24 | Method of speech processing and equipment |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910335969.7A CN110189751A (en) | 2019-04-24 | 2019-04-24 | Method of speech processing and equipment |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN110189751A true CN110189751A (en) | 2019-08-30 |
Family
ID=67715029
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910335969.7A Pending CN110189751A (en) | 2019-04-24 | 2019-04-24 | Method of speech processing and equipment |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110189751A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111062221A (en) * | 2019-12-13 | 2020-04-24 | 北京欧珀通信有限公司 | Data processing method, data processing device, electronic equipment and storage medium |
| CN111143595A (en) * | 2019-12-27 | 2020-05-12 | 上海擎感智能科技有限公司 | Picture management method, system, medium and device based on voice recognition |
| CN111710338A (en) * | 2020-06-28 | 2020-09-25 | 上海优扬新媒信息技术有限公司 | Voice operation playing method and device |
| CN113434670A (en) * | 2021-06-22 | 2021-09-24 | 未鲲(上海)科技服务有限公司 | Method and device for generating dialogistic text, computer equipment and storage medium |
| CN113613068A (en) * | 2021-08-03 | 2021-11-05 | 北京字跳网络技术有限公司 | Video processing method and device, electronic equipment and storage medium |
| CN113791981A (en) * | 2021-09-18 | 2021-12-14 | 平安科技（深圳）有限公司 | Intention operation test method, device, equipment and storage medium |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003223185A (en) * | 2002-01-31 | 2003-08-08 | Nippon Telegr & Teleph Corp <Ntt> | Speech comprehension method and device, speech comprehension program and storage medium stored with speech comprehension program |
| CN106934068A (en) * | 2017-04-10 | 2017-07-07 | 江苏东方金钰智能机器人有限公司 | The method that robot is based on the semantic understanding of environmental context |
| CN107274903A (en) * | 2017-05-26 | 2017-10-20 | 北京搜狗科技发展有限公司 | Text handling method and device, the device for text-processing |
| CN107656921A (en) * | 2017-10-10 | 2018-02-02 | 上海数眼科技发展有限公司 | A kind of short text dependency analysis method based on deep learning |
| CN107967250A (en) * | 2016-10-19 | 2018-04-27 | 中兴通讯股份有限公司 | A kind of information processing method and device |
| CN107992543A (en) * | 2017-11-27 | 2018-05-04 | 上海智臻智能网络科技股份有限公司 | Question and answer exchange method and device, computer equipment and computer-readable recording medium |
| CN108460018A (en) * | 2018-02-28 | 2018-08-28 | 首都师范大学 | A kind of Chinese chapter theme expression power analysis method based on syntax predicate cluster |
| CN109361823A (en) * | 2018-11-01 | 2019-02-19 | 深圳市号互联科技有限公司 | A kind of intelligent interaction mode that voice is mutually converted with text |
| CN109377998A (en) * | 2018-12-11 | 2019-02-22 | 科大讯飞股份有限公司 | A kind of voice interactive method and device |
-
2019
- 2019-04-24 CN CN201910335969.7A patent/CN110189751A/en active Pending
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003223185A (en) * | 2002-01-31 | 2003-08-08 | Nippon Telegr & Teleph Corp <Ntt> | Speech comprehension method and device, speech comprehension program and storage medium stored with speech comprehension program |
| CN107967250A (en) * | 2016-10-19 | 2018-04-27 | 中兴通讯股份有限公司 | A kind of information processing method and device |
| CN106934068A (en) * | 2017-04-10 | 2017-07-07 | 江苏东方金钰智能机器人有限公司 | The method that robot is based on the semantic understanding of environmental context |
| CN107274903A (en) * | 2017-05-26 | 2017-10-20 | 北京搜狗科技发展有限公司 | Text handling method and device, the device for text-processing |
| CN107656921A (en) * | 2017-10-10 | 2018-02-02 | 上海数眼科技发展有限公司 | A kind of short text dependency analysis method based on deep learning |
| CN107992543A (en) * | 2017-11-27 | 2018-05-04 | 上海智臻智能网络科技股份有限公司 | Question and answer exchange method and device, computer equipment and computer-readable recording medium |
| CN108460018A (en) * | 2018-02-28 | 2018-08-28 | 首都师范大学 | A kind of Chinese chapter theme expression power analysis method based on syntax predicate cluster |
| CN109361823A (en) * | 2018-11-01 | 2019-02-19 | 深圳市号互联科技有限公司 | A kind of intelligent interaction mode that voice is mutually converted with text |
| CN109377998A (en) * | 2018-12-11 | 2019-02-22 | 科大讯飞股份有限公司 | A kind of voice interactive method and device |
Non-Patent Citations (1)
| Title |
|---|
| 中国图书馆学会编译出版委员会: "《图书馆学研究论文集》", 30 June 1996, 北京：书目文献出版社 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111062221A (en) * | 2019-12-13 | 2020-04-24 | 北京欧珀通信有限公司 | Data processing method, data processing device, electronic equipment and storage medium |
| CN111143595A (en) * | 2019-12-27 | 2020-05-12 | 上海擎感智能科技有限公司 | Picture management method, system, medium and device based on voice recognition |
| CN111710338A (en) * | 2020-06-28 | 2020-09-25 | 上海优扬新媒信息技术有限公司 | Voice operation playing method and device |
| CN113434670A (en) * | 2021-06-22 | 2021-09-24 | 未鲲(上海)科技服务有限公司 | Method and device for generating dialogistic text, computer equipment and storage medium |
| WO2022267174A1 (en) * | 2021-06-22 | 2022-12-29 | 未鲲(上海)科技服务有限公司 | Script text generating method and apparatus, computer device, and storage medium |
| CN113613068A (en) * | 2021-08-03 | 2021-11-05 | 北京字跳网络技术有限公司 | Video processing method and device, electronic equipment and storage medium |
| CN113791981A (en) * | 2021-09-18 | 2021-12-14 | 平安科技（深圳）有限公司 | Intention operation test method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110189751A (en) | Method of speech processing and equipment | |
| CN110377716A (en) | Exchange method, device and the computer readable storage medium of dialogue | |
| CN108804414A (en) | Text modification method, device, smart machine and readable storage medium storing program for executing | |
| CN110222182B (en) | Statement classification method and related equipment | |
| CN108847241A (en) | It is method, electronic equipment and the storage medium of text by meeting speech recognition | |
| CN111125354A (en) | Text classification method and device | |
| CN108447471A (en) | Audio recognition method and speech recognition equipment | |
| CN110262273A (en) | A kind of home equipment control method, device, storage medium and smart home system | |
| CN108228704A (en) | Identify method and device, the equipment of Risk Content | |
| CN107402913A (en) | The determination method and apparatus of antecedent | |
| CN111445898B (en) | Language identification method and device, electronic equipment and storage medium | |
| CN108682420A (en) | A kind of voice and video telephone accent recognition method and terminal device | |
| CN108416032A (en) | A kind of file classification method, device and storage medium | |
| CN111414746A (en) | Matching statement determination method, device, equipment and storage medium | |
| CN111191463A (en) | Emotion analysis method and device, electronic equipment and storage medium | |
| CN109859747A (en) | Voice interactive method, equipment and storage medium | |
| CN113486170A (en) | Natural language processing method, device, equipment and medium based on man-machine interaction | |
| CN108268443A (en) | It determines the transfer of topic point and obtains the method, apparatus for replying text | |
| CN110020429A (en) | Method for recognizing semantics and equipment | |
| CN115292492A (en) | Method, device and equipment for training intention classification model and storage medium | |
| CN112863518B (en) | Method and device for recognizing voice data subject | |
| Khasanova et al. | Developing a production system for Purpose of Call detection in business phone conversations | |
| CN110188330B (en) | Method and device for determining similar text information, electronic equipment and storage medium | |
| CN114239602A (en) | Session method, apparatus and computer program product | |
| CN110276001B (en) | Checking page identification method and device, computing equipment and medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190830 |
|
| RJ01 | Rejection of invention patent application after publication |