CN108595412A - Correction processing method and device, computer equipment and readable medium - Google Patents
Correction processing method and device, computer equipment and readable medium Download PDFInfo
- Publication number
- CN108595412A CN108595412A CN201810225708.5A CN201810225708A CN108595412A CN 108595412 A CN108595412 A CN 108595412A CN 201810225708 A CN201810225708 A CN 201810225708A CN 108595412 A CN108595412 A CN 108595412A
- Authority
- CN
- China
- Prior art keywords
- read statement
- raw tone
- service
- error correction
- environment
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/205—Parsing
- G06F40/211—Syntactic parsing, e.g. based on context-free grammar [CFG] or unification grammars
-
- 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
- G10L15/26—Speech to text systems
Abstract
A kind of correction processing method and device of present invention offer, computer equipment and readable medium.Its method includes：Receive the raw tone read statement for presetting the user in environment；Attempt to carry out service according to raw tone read statement to recall processing；If can not recall respective service according to raw tone read statement, according to the error correction map table excavated in advance in default environment, correction process is carried out to raw tone read statement.Technical scheme of the present invention can be based on error correction map table, realize to carrying out correction process to raw tone read statement, and then can effectively improve service recall rate, enhance the usage experience of user.And in the present invention, for each use environment, it can realize that correction process, use are very flexible based on the error correction map table under the environment.
Description
【Technical field】
The present invention relates to computer application technologies more particularly to a kind of correction processing method and device, computer to set
Standby and readable medium.
【Background technology】
With the fast development of intelligent terminal the relevant technologies, user can directly pass through voice and intelligent terminal
It interacts, the thorough liberation both hands of user greatly improve the usage experience of user.For example, the relevant technologies are based on, it can
It is various similar to intelligent terminals such as children-story machine, intelligent sound boxes to develop.
Existing intelligent terminal at work, can receive the voice read statement i.e. Query sentences of user, then
Automatic speech recognition (Automatic Speech Recognition are carried out to the Query sentences of user；ASR), identified
Text results afterwards.It is then based on text results and carries out recalling for respective service.If recalling respective service, phase is being pushed to user
It should service；If recalling less than respective service, any response is not done.The user that existing intelligent terminal is supported inputs language
Sound can be Chinese either English or other language.But for each language, a kind of pronunciation of standard is only supported.If such as
When language input by user is Chinese, intelligent terminal usually only supports that the voice of input is the Mandarin Chinese of standard.But
It is in practical application, intelligent terminal in the market may sell each place, and the user group used is also very multiple
It is miscellaneous, such as may have and serve the user with accent, it is also possible to serve the user of the included intonation sprouted；It is possible that serving
User from clear articulation and a mellow and full tone northeast, it is also possible to serve the user for the southwest that tongue portion is stuck up from flat tongue.
The complexity of user group based on intelligent terminal, the voice that intelligent terminal has the user received are defeated
The pronunciation for entering sentence is nonstandard, and the prior art lacks the error correction to voice read statement, so as to cause intelligent terminal
Service recall rate it is smaller.
【Invention content】
The present invention provides a kind of correction processing method and device, computer equipment and readable mediums, for improving intelligence
The service recall rate of terminal device.
The present invention provides a kind of correction processing method, the method includes：
Receive the raw tone read statement for presetting the user in environment；
Attempt to carry out service according to the raw tone read statement to recall processing；
If can not recall respective service according to the raw tone read statement, according to advance in the default environment
The error correction map table of excavation carries out correction process to the raw tone read statement.
Still optionally further, in method as described above, attempt to carry out service according to the raw tone read statement to call together
Processing is returned, is specifically included：
Speech recognition is carried out to the raw tone read statement, obtains corresponding original character sentence；
According to the original character sentence, attempt to carry out service from preset set of service to recall processing.
Still optionally further, in method as described above, according to the error correction map table excavated in advance in the default environment,
Correction process is carried out to the raw tone read statement, is specifically included：
According to the error correction map table excavated in advance in the default environment, the raw tone read statement is corresponded to
The original character sentence carry out correction process, obtain the corresponding target text sentence of the raw tone read statement.
Still optionally further, in method as described above, according to the error correction map table excavated in advance in the default environment,
After carrying out correction process to the raw tone read statement, the method further includes：
According to the target text sentence, attempt to carry out service from the set of service to recall processing.
Still optionally further, in method as described above, the raw tone read statement for presetting the user in environment is received,
It specifically includes：
The raw tone read statement for the user that the intelligent terminal in the default environment is sent is received, it is described
The raw tone read statement of user is intelligent terminal acquisition.
Still optionally further, in method as described above, according to the error correction map table excavated in advance in the default environment,
Before carrying out correction process to the raw tone read statement, the method further includes：
It acquires in the default environment, either all voice inputs of preset times in the preset acquisition time period
At least one correct voice read statement of respective service can be recalled in sentence and all voice read statements；
It excavates from all voice read statements and is less than in advance with the diversity factor of each correct voice read statement
If being more than in advance with the number of the correct voice read statement occurred jointly in diversity factor threshold value, and/or the collection period
If the garbled voice read statement of frequency threshold value；
Error correction map will be established between each correct voice read statement and the corresponding garbled voice read statement
Relationship obtains the error correction map table.
The present invention provides a kind of error correcting handling arrangement, and described device includes：
Receiving module, the raw tone read statement for receiving the user in default environment；
Processing module is recalled, processing is recalled for attempting to carry out service according to the raw tone read statement；
Correction module, if when for respective service can not to be recalled according to the raw tone read statement, according to described
The error correction map table excavated in advance in default environment carries out correction process to the raw tone read statement.
Still optionally further, described to recall processing module in device as described above, it is specifically used for：
Speech recognition is carried out to the raw tone read statement, obtains corresponding original character sentence；
According to the original character sentence, attempt to carry out service from preset set of service to recall processing.
Still optionally further, in device as described above, the correction module is specifically used for according in the default environment
The error correction map table excavated in advance, the original character sentence corresponding to the raw tone read statement carry out error correction
Processing, obtains the corresponding target text sentence of the raw tone read statement.
Still optionally further, described to recall processing module in device as described above, it is additionally operable to according to the target text
Sentence, attempts to carry out service from the set of service to recall processing.
Still optionally further, in device as described above, the receiving module is specifically used for receiving in the default environment
Intelligent terminal send the user raw tone read statement, the raw tone read statement of the user is institute
State intelligent terminal acquisition.
Still optionally further, in device as described above, further include：
Acquisition module, for acquiring in the default environment, either preset times in the preset acquisition time period
At least one correct of respective service can be recalled in all voice read statements and all voice read statements
Voice read statement；
Module is excavated, for being excavated from all voice read statements and each correct voice read statement
Diversity factor is less than the common appearance with the correct voice read statement in default diversity factor threshold value, and/or the collection period
Number be more than preset times threshold value garbled voice read statement；
Module is established, being used for will be between each correct voice read statement and the corresponding garbled voice read statement
Error correction map relationship is established, the error correction map table is obtained.
Still optionally further, in device as described above, the error correcting handling arrangement is arranged in terminal device or high in the clouds
In server.
The present invention also provides a kind of computer equipment, the equipment includes：
One or more processors；
Memory, for storing one or more programs；
When one or more of programs are executed by one or more of processors so that one or more of processing
Device realizes correction processing method as described above.
The present invention also provides a kind of computer-readable mediums, are stored thereon with computer program, which is held by processor
Correction processing method as described above is realized when row.
Correction processing method and device, the computer equipment and readable medium of the present invention, is preset by reception in environment
The raw tone read statement of user；Attempt to carry out service according to raw tone read statement to recall processing；If according to original language
When sound read statement can not recall respective service, according to the error correction map table excavated in advance in default environment, to raw tone
Read statement carries out correction process.Technical scheme of the present invention can be based on error correction map table, realize to being inputted to raw tone
Sentence carries out correction process, and then can effectively improve service recall rate, enhances the usage experience of user.And in the present invention,
For each use environment, it can realize that correction process, use are very flexible based on the error correction map table under the environment.
【Description of the drawings】
Fig. 1 is the flow chart of the correction processing method embodiment one of the present invention.
Fig. 2 is the flow chart of the correction processing method embodiment two of the present invention.
Fig. 3 is the structure chart of the error correcting handling arrangement embodiment one of the present invention.
Fig. 4 is the structure chart of the error correcting handling arrangement embodiment two of the present invention.
Fig. 5 is the structure chart of the computer equipment embodiment of the present invention.
Fig. 6 is a kind of exemplary plot of computer equipment provided by the invention.
【Specific implementation mode】
To make the objectives, technical solutions, and advantages of the present invention clearer, right in the following with reference to the drawings and specific embodiments
The present invention is described in detail.
Fig. 1 is the flow chart of the correction processing method embodiment one of the present invention.As shown in Figure 1, at the error correction of the present embodiment
Reason method, can specifically include following steps：
100, the raw tone read statement for presetting the user in environment is received；
101, attempt to carry out service according to raw tone read statement to recall processing；
If 102, respective service can not be recalled according to raw tone read statement, excavated in advance according in default environment
Error correction map table, to raw tone read statement carry out correction process.
The executive agent of the error correction method of the present embodiment is error correcting handling arrangement, which can be arranged in intelligence
In energy terminal device.I.e. the usage scenario of the correction processing method of the present embodiment is in intelligent terminal.At this point, intelligent terminal
Equipment can receive the intelligent terminal of the voice Query of user for such as intelligent sound box, intelligent children-story machine, and
The intelligent terminal can also provide service to the user based on the voice Query of user.
Or the error correcting handling arrangement of the present embodiment can also be arranged in the cloud being connect with intelligent terminal wireless telecommunications
It holds in server.That is the usage scenario of the correction processing method of the present embodiment is beyond the clouds in server.At this point, intelligent terminal
The voice Query that the user in the use environment of intelligent terminal can only be received, then reports to cloud server
The voice Query of acquisition passes through the intelligence by the voice Query for the user that cloud server is reported according to the intelligent terminal
Energy terminal device provides service to the user.
Since in practical application, intelligent terminal can be sold to different places, there is difference in different places
Language environment, such as intelligent terminal may by sell to one it is flat stick up tongue regardless of family in, it is also possible to sold
A kindergarten is sold for, the child also to pronounce indistinctly that can speak for a group provides respective service.For each environment, all
There are its language characteristics, the present embodiment to need intelligent terminal can be to the ring to improve the service recall rate in the environment
Voice read statement in border carries out error correction, is recalled with obtaining more services according to the voice read statement after error correction, in turn
Improve service recall rate.
For example, when if the error correcting handling arrangement of the present embodiment is located in intelligent terminal, at this point, the step 100 is specific
It can be the raw tone read statement that the user in environment is preset in intelligent terminal acquisition.And then by intelligent terminal root
Attempt to carry out service according to raw tone read statement to recall processing；Then intelligent terminal judges whether to recall corresponding clothes
Business, if recalling, directly provides a user respective service.If otherwise not recalling, intelligent terminal is according to default at this time
The error correction map table excavated in advance in environment carries out correction process to raw tone read statement, can subsequently be based on entangling in this way
Voice read statement after mistake carries out service and recalls.
And if the error correcting handling arrangement of the present embodiment is when being located in cloud server, at this point, the step 100 is specifically as follows
Cloud server receives intelligent terminal and acquires and the raw tone read statement of user in the default environment that reports.By cloud
End server attempts to carry out service to recall processing according to raw tone read statement；If can not be called together according to raw tone read statement
When returning to respective service, according to the error correction map table excavated in advance in default environment, error correction is carried out to raw tone read statement
Processing.Then cloud server judges whether to recall respective service, if recalling, is directly called together to intelligent terminal push
The service returned, so that intelligent terminal provides a user respective service.And if do not recall, cloud server is according to pre- at this time
If the error correction map table excavated in advance in environment, correction process is carried out to raw tone read statement, it so subsequently can be with high in the clouds
Server can carry out service based on the voice read statement after error correction and recall, and after recalling, and be pushed to intelligent terminal
The service recalled, so that intelligent terminal provides a user the service recalled.
No matter under which kind of scene of the present embodiment, respective service can not recalled according to raw tone read statement
When, it is required to, according to the error correction map table excavated in advance in environment is preset, correction process be carried out to raw tone read statement.
May include some common error correction maps passes in the default environment in the error correction map table wherein excavated in advance
System.Such as the intelligent terminal, when having just enter into a certain environment, intelligent terminal needs constantly to adapt to the environment, receiving should
One section of acquisition time period of user in environment or the voice input of certain number, can excavate one in the environment
Error correction map relationship between a little mistake Quey and correct Query, forms error correction map table.
Such as front and back input two Query sentences of A and B of user in certain environment, if user inputs after A, when presetting
Between do not provide related service in the period, then it represents that A input by user is recalled.The preset period of time can be that response services most
The big time cycle.And and then if user inputs after B, intelligent terminal provides corresponding service, and user's short time
Do not continue to Query inside, then it represents that user for B result when meet.Based on the above situation, it can be inferred that A may be
The Query of mistake, B may be correct Query.It is then based on the diversity factor of A and B and number that A and B occur jointly, presses
It is more than less than sentence occurrence number in this presets environment as default diversity factor threshold value and/or A-B according to the diversity factor of A and B
Preset times threshold value can then filter out the error correction pairing that A-B is " mistake-is correct " in a speech recognition.Wherein input A
It is smaller with the diversity factor of B, indicate that A and B is bigger as the probability of error correction pair, such as after the A sentences that have an accent of user's input tape,
After not recalling respective service, it is appreciated that certain word pronunciations in A are dialect, and it is B input by sentence immediately to change A sentences, then
Respective service is recalled, at this point, A and B is a pair of of error correction pair, the diversity factor of A and B are smaller.In addition, in the data of acquisition, A and
The number that B occurs together is more, then it represents that A and B is also bigger as the probability of error correction pair.If A can be the user in the environment
The sentence of dialect input is carried, and B is the sentence of the standard mandarin input of user in the environment.Specifically, user often exists
After the read statement A for carrying dialect, when not recalling respective service, read statement A is adjusted to standard mandarin version in time
This read statement B, recalls respective service, in this case, A and B just belong to common appearance.That is, the present embodiment
Occur being that there is certain scene jointly, it is necessary to be to occur together in a very short preset period of time, and there are one call together
Respective service is returned to, another does not recall respective service.
The diversity factor of two sentences of A and B in the present embodiment is equal to the smallest edit distance of A and B phonetics/(long in A and B
The phonetic character of Query).Wherein molecule is the editing distance of two pinyin character strings of A and B, refers to utilizing character manipulation, word
Symbol string A is converted into the required minimal action numbers of character string B.Wherein, character manipulation includes deleting a character, being inserted into a word
Symbol, three kinds of the character operation of modification one.Smallest edit distance can use the algorithm of Dynamic Programming to obtain.Denominator can be expressed as
Max (len (phonetic of A), len (phonetic of B)), i.e., the length of length most elder in the phonetic of A, B.
For example, in an environment of actual scene, user has first said " I will listen woman's story of wandering ", intelligent terminal
Equipment returns " sorry, not find the story ", then it represents that intelligent terminal is not recalled；If user has subsequently added " I
Listen the story of the Cowherd and the Weaving Maid ", intelligent terminal returns correct story, then it represents that equipment is recalled.And it sends out by analysis
It is existing：A sentences " I will listen woman's story of wandering " phonetic is " woyaotingliulangzhinvdegushi ", and B sentences are
“woyaotingniulangzhinvdegushi”.The diversity factor of above-mentioned A and B is calculated as 1/28, such as default diversity factor threshold value
When being 0.2,1/28 well below 0.2 (reference value), then A is an error correction of B, and A and B are a pair of of error correction pair.If in history number
According to kind, occurrence number is 3 times together by A and B, if predetermined threshold value number is 2 times, can also verify the error correction that A is B, A and B
For a pair of of error correction pair.
In the manner described above, the one section of acquisition time that can have just been begun to use according to the intelligent terminal in default environment
Data in period, or just begun to use the data of certain preset times, excavate all error correction pair preset in environment, shape
At error correction map table.The acquisition time period of the present embodiment can be arranged according to actual demand, such as can be 1 week, 1 month
Or the time cycle of other length.Preset times can be 500 times, 1000 times or other number numerical value, the present embodiment
The number of the voice read statement of acquisition user may be used using certain preset times to indicate.
May include several in the error correction map table excavated by aforesaid way to error correction map relationship, per a pile error correction map
Relationship includes a correct voice read statement and a corresponding wrong voice read statement.And different error correction
May include identical correct voice read statement in mapping relations.Such as user successively inputs A, B and C sentence, if A and B
Sentence does not recall service, and the service of recalling of C sentences, and A and B, A and C be satisfied by diversity factor be less than default diversity factor threshold value,
And/or the number occurred jointly is more than preset times threshold value, then it is also a pair of of error correction that A and B, which is a pair of of error correction map relationship, A and C,
Mapping relations.In this way, when according to raw tone read statement respective service can not be recalled, it can be according to pre- in default environment
The error correction map table first excavated, judgement are originally inputted in some the error correction map relationship whether sentence hits in error correction map table
Garbled voice read statement, if hit, using corresponding error correction map Relation acquisition, the garbled voice read statement is corresponding
Correct voice read statement, to be entangled to the voice read statement of mistake input by user using the correct voice read statement
It is wrong.Service subsequently can be carried out according to the correct voice read statement to recall, so as to effectively improve service recall rate.
The correction processing method of the present embodiment presets the raw tone read statement of the user in environment by reception；Root
Attempt to carry out service according to raw tone read statement to recall processing；If corresponding clothes can not be recalled according to raw tone read statement
When business, according to the error correction map table excavated in advance in default environment, correction process is carried out to raw tone read statement.This implementation
The technical solution of example can be based on error correction map table, realize to carrying out correction process to raw tone read statement, and then can
Service recall rate is effectively improved, the usage experience of user is enhanced.And in the present embodiment, for each use environment,
To realize that correction process, use are very flexible based on the error correction map table under the environment.
Fig. 2 is the flow chart of the correction processing method embodiment two of the present invention.As shown in Fig. 2, at the error correction of the present embodiment
Reason method, on the basis of the technical solution of above-described embodiment, by taking error correcting handling arrangement setting beyond the clouds server as an example, to retouch
State technical scheme of the present invention.The correction processing method of the present embodiment, can specifically include following steps：
200, the intelligent terminal in default environment acquires in the default environment, in the preset acquisition time period
Each voice read statement, and report cloud server；
201, cloud server receives all of intelligent terminal being located in default environment and reports, and acquires and preset ring
It can be recalled in all voice read statements and all voice read statements in border, in the preset acquisition time period
At least one correct voice read statement of respective service；
202, cloud server is excavated corresponding with each correct voice read statement from all voice read statements
Garbled voice read statement；
The excavation principle of the present embodiment can be to excavate to be less than default difference with the diversity factor of each correct voice read statement
It spends big with the number of correct voice read statement occurred jointly in the garbled voice read statement, and/or collection period of threshold value
In the garbled voice read statement of preset times threshold value, the record of above-described embodiment can be referred in detail, details are not described herein.
203, cloud server will be established between each correct voice read statement and corresponding garbled voice read statement and be entangled
Wrong mapping relations obtain error correction map table；
Above-mentioned steps 200-203 can be understood as establishing the off-line operation of the error correction map table in the default environment.Subsequently
Step 204-211 can be according to the error correction map table of foundation, and the voice read statement to not recalling respective service entangles
Mistake, with the recall rate for the service of improving.
204, cloud server receives the original of the user for acquiring and reporting positioned at the intelligent terminal preset in environment
Voice read statement；
205, cloud server carries out speech recognition to raw tone read statement, obtains corresponding original character sentence；
206, cloud server is according to original character sentence, from attempting to carry out in preset set of service from service recalls
Reason；
207, cloud server judges whether to recall respective service, if not recalling, executes step 208；Otherwise it executes
Step 209；
208, cloud server is according to the error correction map table excavated in advance in default environment, to raw tone read statement pair
The original character sentence answered carries out correction process, obtains the corresponding target text sentence of raw tone read statement；Execute step
210；
209, cloud server is according to original character sentence, attempts to carry out service from set of service to recall processing；It executes
Step 211；
210, cloud server is according to target text sentence, attempts to carry out service from set of service to recall processing；It executes
Step 211；
211, when cloud server recalls respective service, the service recalled is pushed to intelligent terminal, for intelligence
Terminal device provides the service recalled to the user, terminates.
The set of service of the present embodiment is the set for all services being capable of providing, such as is youngster for intelligent terminal
When virgin Story machine, the voice data of numerous children stories can be stored in the set of service, in order to provide children stories
Service.For example, when child by children-story machine ask " JackKen " when, cloud server can with the set of service,
The audio data of " JackKen " is therefrom obtained, and is pushed to children stories collection, so that children-story machine is " small to child's offer
Horse crosses the river " audio service.
The correction processing method of the present embodiment is that server realizes technical scheme of the present invention beyond the clouds, implemented in detail
Journey can also refer to the related of above-mentioned embodiment illustrated in fig. 1 and record, and details are not described herein.
The correction processing method of the present embodiment can be based on error correction map table, realization pair by using above-mentioned technical proposal
Correction process is carried out to raw tone read statement, and then service recall rate can be effectively improved, enhance user uses body
It tests.And in the present embodiment, for each use environment, it can be realized at error correction based on the error correction map table under the environment
Reason, use are very flexible.
Fig. 3 is the structure chart of the error correcting handling arrangement embodiment one of the present invention.As shown in figure 3, at the error correction of the present embodiment
Device is managed, can specifically include：
Receiving module 10 is used to receive the raw tone read statement of the user in default environment；
Processing module 11 is recalled to attempt to carry out service for the raw tone read statement that is received according to receiving module 10 to call together
Return processing；
If correction module 12 can not recall respective service for recalling processing module 11 according to raw tone read statement
When, according to the error correction map table excavated in advance in default environment, the raw tone read statement that receiving module 10 receives is carried out
Correction process.
The error correcting handling arrangement of the present embodiment realizes the realization principle and technology of correction process by using above-mentioned module
Effect is identical as the realization of above-mentioned related method embodiment, can refer to the record of above-mentioned related method embodiment in detail, herein
It repeats no more.
Fig. 4 is the structure chart of the error correcting handling arrangement embodiment two of the present invention.As shown in figure 4, at the error correction of the present embodiment
Device is managed, on the basis of the technical solution of above-mentioned embodiment illustrated in fig. 3, further introduces the technology of the present invention in further detail
Scheme.
In the error correcting handling arrangement of the present embodiment, recalls processing module 11 and be specifically used for：
Speech recognition is carried out to raw tone read statement, obtains corresponding original character sentence；
According to original character sentence, attempt to carry out service from preset set of service to recall processing.
Still optionally further, in the error correcting handling arrangement of the present embodiment, correction module 12 is specifically used for according to default environment
In the error correction map table that excavates in advance, the corresponding original character sentence of raw tone read statement that receiving module 10 is received into
Row correction process obtains the corresponding target text sentence of raw tone read statement.
Still optionally further, it in the error correcting handling arrangement of the present embodiment, recalls processing module 11 and is additionally operable to according to target text
Word sentence, attempts to carry out service from set of service to recall processing.
Still optionally further, in the error correcting handling arrangement of the present embodiment, receiving module 10, which is specifically used for receiving, presets environment
In the raw tone read statement of user that sends of intelligent terminal, the raw tone read statement of user is intelligent terminal
Equipment acquisition.
Still optionally further, as shown in figure 4, in the error correcting handling arrangement of the present embodiment, further include：
Acquisition module 13 is used to acquire in default environment, either preset times is all in the preset acquisition time period
Voice read statement and all voice read statements in can recall at least one correct voice input of respective service
Sentence；
It excavates in all voice read statements that module 14 is used to acquire from acquisition module 13 and excavates and each correct voice
The diversity factor of read statement is less than in default diversity factor threshold value, and/or collection period to be gone out jointly with correct voice read statement
Existing number is more than the garbled voice read statement of preset times threshold value；
It is defeated for each correct voice read statement that module 14 is excavated and corresponding garbled voice will to be excavated to establish module 15
Enter and establish error correction map relationship between sentence, obtains error correction map table.
Accordingly, if correction module 12 can not recall phase for recalling processing module 11 according to raw tone read statement
When should service, according to the error correction map table excavated in advance in the default environment for establishing the foundation of module 15, receiving module 10 is received
The corresponding original character sentence of raw tone read statement carry out correction process, obtain the corresponding mesh of raw tone read statement
Mark word sentence.
Still optionally further, the error correcting handling arrangement of the present embodiment can be arranged in terminal device or cloud server
In.
The error correcting handling arrangement of the present embodiment realizes the realization principle and technology of correction process by using above-mentioned module
Effect is identical as the realization of above-mentioned related method embodiment, can refer to the record of above-mentioned related method embodiment in detail, herein
It repeats no more.
Fig. 5 is the structure chart of the computer equipment embodiment of the present invention.As shown in figure 5, the computer equipment of the present embodiment,
Including：One or more processors 30 and memory 40, memory 40 work as memory for storing one or more programs
The one or more programs stored in 40 are executed by one or more processors 30 so that one or more processors 30 are realized such as
The correction processing method of figure 1 above-embodiment illustrated in fig. 2.In embodiment illustrated in fig. 5 for including multiple processors 30.Such as
The computer equipment of the present embodiment is specifically as follows intelligent terminal, or may be cloud server equipment.
For example, Fig. 6 is a kind of exemplary plot of computer equipment provided by the invention.Fig. 6 is shown suitable for being used for realizing this
The block diagram of the exemplary computer device 12a of invention embodiment.The computer equipment 12a that Fig. 6 is shown is only an example,
Any restrictions should not be brought to the function and use scope of the embodiment of the present invention.
As shown in fig. 6, computer equipment 12a is showed in the form of universal computing device.The component of computer equipment 12a can
To include but not limited to：One or more processor 16a, system storage 28a, connection different system component (including system
Memory 28a and processor 16a) bus 18a.
Bus 18a indicates one or more in a few class bus structures, including memory bus or Memory Controller,
Peripheral bus, graphics acceleration port, processor or the local bus using the arbitrary bus structures in a variety of bus structures.It lifts
For example, these architectures include but not limited to industry standard architecture (ISA) bus, microchannel architecture (MAC)
Bus, enhanced isa bus, Video Electronics Standards Association (VESA) local bus and peripheral component interconnection (PCI) bus.
Computer equipment 12a typically comprises a variety of computer system readable media.These media can be it is any can
The usable medium accessed by computer equipment 12a, including volatile and non-volatile media, moveable and immovable Jie
Matter.
System storage 28a may include the computer system readable media of form of volatile memory, such as deposit at random
Access to memory (RAM) 30a and/or cache memory 32a.Computer equipment 12a may further include it is other it is removable/
Immovable, volatile/non-volatile computer system storage medium.Only as an example, storage system 34a can be used for reading
Write immovable, non-volatile magnetic media (Fig. 6 do not show, commonly referred to as " hard disk drive ").Although being not shown in Fig. 6,
It can provide for the disc driver to moving non-volatile magnetic disk (such as " floppy disk ") read-write, and to removable non-easy
The CD drive that the property lost CD (such as CD-ROM, DVD-ROM or other optical mediums) is read and write.In these cases, each
Driver can be connected by one or more data media interfaces with bus 18a.System storage 28a may include at least
There is one group of (for example, at least one) program module, these program modules to be configured to hold for one program product, the program product
The function of the above-mentioned each embodiments of Fig. 1-Fig. 4 of the row present invention.
Program with one group of (at least one) program module 42a/utility 40a can be stored in such as system and deposit
In reservoir 28a, such program module 42a include --- but being not limited to --- operating system, one or more application program,
Other program modules and program data may include the reality of network environment in each or certain combination in these examples
It is existing.Program module 42a usually executes the function and/or method in above-mentioned each embodiments of Fig. 1-Fig. 4 described in the invention.
Computer equipment 12a can also be with one or more external equipment 14a (such as keyboard, sensing equipment, display
24a etc.) communication, the equipment interacted with computer equipment 12a communication can be also enabled a user to one or more, and/or
(such as network interface card is adjusted with any equipment that computer equipment 12a communicated with one or more of the other computing device is enable
Modulator-demodulator etc.) communication.This communication can be carried out by input/output (I/O) interface 22a.Also, computer equipment
12a can also by network adapter 20a and one or more network (such as LAN (LAN), wide area network (WAN) and/or
Public network, such as internet) communication.As shown, network adapter 20a by bus 18a and computer equipment 12a its
Its module communicates.It should be understood that although not shown in the drawings, other hardware and/or software can be used in conjunction with computer equipment 12a
Module, including but not limited to：Microcode, device driver, redundant processor, external disk drive array, RAID system, tape
Driver and data backup storage system etc..
Processor 16a is stored in program in system storage 28a by operation, to perform various functions application and
Data processing, such as realize correction processing method shown in above-described embodiment.
The present invention also provides a kind of computer-readable mediums, are stored thereon with computer program, which is held by processor
The correction processing method as shown in above-described embodiment is realized when row.
The computer-readable medium of the present embodiment may include in the system storage 28a in above-mentioned embodiment illustrated in fig. 6
RAM30a, and/or cache memory 32a, and/or storage system 34a.
With the development of science and technology, the route of transmission of computer program is no longer limited by tangible medium, it can also be directly from net
Network is downloaded, or is obtained using other modes.Therefore, the computer-readable medium in the present embodiment may include not only tangible
Medium can also include invisible medium.
The arbitrary combination of one or more computer-readable media may be used in the computer-readable medium of the present embodiment.
Computer-readable medium can be computer-readable signal media or computer readable storage medium.Computer-readable storage medium
Matter for example may be-but not limited to-system, device or the device of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, or
The arbitrary above combination of person.The more specific example (non exhaustive list) of computer readable storage medium includes：There are one tools
Or the electrical connections of multiple conducting wires, portable computer diskette, hard disk, random access memory (RAM), read-only memory (ROM),
Erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light
Memory device, magnetic memory device or above-mentioned any appropriate combination.In this document, computer readable storage medium can
With to be any include or the tangible medium of storage program, the program can be commanded execution system, device or device use or
Person is in connection.
Computer-readable signal media may include in a base band or as the data-signal that a carrier wave part is propagated,
Wherein carry computer-readable program code.Diversified forms may be used in the data-signal of this propagation, including --- but
It is not limited to --- electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be
Any computer-readable medium other than computer readable storage medium, which can send, propagate or
Transmission for by instruction execution system, device either device use or program in connection.
The program code for including on computer-readable medium can transmit with any suitable medium, including --- but it is unlimited
In --- wireless, electric wire, optical cable, RF etc. or above-mentioned any appropriate combination.
It can be write with one or more programming languages or combinations thereof for executing the computer that operates of the present invention
Program code, described program design language include object oriented program language-such as Java, Smalltalk, C++,
Further include conventional procedural programming language-such as " C " language or similar programming language.Program code can be with
It fully executes, partly execute on the user computer on the user computer, being executed as an independent software package, portion
Divide and partly executes or executed on a remote computer or server completely on the remote computer on the user computer.
Be related in the situation of remote computer, remote computer can pass through the network of any kind --- including LAN (LAN) or
Wide area network (WAN)-be connected to subscriber computer, or, it may be connected to outer computer (such as carried using Internet service
It is connected by internet for quotient).
In several embodiments provided by the present invention, it should be understood that disclosed system, device and method can be with
It realizes by another way.For example, the apparatus embodiments described above are merely exemplary, for example, the unit
It divides, only a kind of division of logic function, formula that in actual implementation, there may be another division manner.
The unit illustrated as separating component may or may not be physically separated, aobvious as unit
The component shown may or may not be physical unit, you can be located at a place, or may be distributed over multiple
In network element.Some or all of unit therein can be selected according to the actual needs to realize the mesh of this embodiment scheme
's.
In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, it can also
It is that each unit physically exists alone, it can also be during two or more units be integrated in one unit.Above-mentioned integrated list
The form that hardware had both may be used in member is realized, can also be realized in the form of hardware adds SFU software functional unit.
The above-mentioned integrated unit being realized in the form of SFU software functional unit can be stored in one and computer-readable deposit
In storage media.Above-mentioned SFU software functional unit is stored in a storage medium, including some instructions are used so that a computer
It is each that equipment (can be personal computer, server or the network equipment etc.) or processor (processor) execute the present invention
The part steps of embodiment the method.And storage medium above-mentioned includes：USB flash disk, mobile hard disk, read-only memory (Read-
Only Memory, ROM), random access memory (Random Access Memory, RAM), magnetic disc or CD etc. it is various
The medium of program code can be stored.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
With within principle, any modification, equivalent substitution, improvement and etc. done should be included within the scope of protection of the invention god.
Claims (15)
1. a kind of correction processing method, which is characterized in that the method includes：
Receive the raw tone read statement for presetting the user in environment；
Attempt to carry out service according to the raw tone read statement to recall processing；
If can not recall respective service according to the raw tone read statement, excavated in advance according in the default environment
Error correction map table, to the raw tone read statement carry out correction process.
2. according to the method described in claim 1, it is characterized in that, attempting to be serviced according to the raw tone read statement
Processing is recalled, is specifically included：
Speech recognition is carried out to the raw tone read statement, obtains corresponding original character sentence；
According to the original character sentence, attempt to carry out service from preset set of service to recall processing.
3. according to the method described in claim 2, it is characterized in that, according to the error correction map excavated in advance in the default environment
Table carries out correction process to the raw tone read statement, specifically includes：
According to the error correction map table excavated in advance in the default environment, institute corresponding to the raw tone read statement
It states original character sentence and carries out correction process, obtain the corresponding target text sentence of the raw tone read statement.
4. according to the method described in claim 3, it is characterized in that, according to the error correction map excavated in advance in the default environment
Table, after carrying out correction process to the raw tone read statement, the method further includes：
According to the target text sentence, attempt to carry out service from the set of service to recall processing.
5. method described in claim 1, which is characterized in that the raw tone read statement for presetting the user in environment is received,
It specifically includes：
Receive the raw tone read statement for the user that the intelligent terminal in the default environment is sent, the user
Raw tone read statement be the intelligent terminal acquisition.
6. according to any methods of claim 1-5, which is characterized in that entangled according to what is excavated in advance in the default environment
Wrong mapping table, before carrying out correction process to the raw tone read statement, the method further includes：
It acquires in the default environment, all voice read statements of interior either preset times of preset acquisition time period
And at least one correct voice read statement of respective service can be recalled in all voice read statements；
It is poor less than default with the diversity factor of each correct voice read statement to be excavated from all voice read statements
It is more than with the number of the correct voice read statement occurred jointly default time in different degree threshold value, and/or the collection period
The garbled voice read statement of number threshold value；
Error correction map relationship will be established between each correct voice read statement and the corresponding garbled voice read statement,
Obtain the error correction map table.
7. a kind of error correcting handling arrangement, which is characterized in that described device includes：
Receiving module, the raw tone read statement for receiving the user in default environment；
Processing module is recalled, processing is recalled for attempting to carry out service according to the raw tone read statement；
Correction module, if when for respective service can not to be recalled according to the raw tone read statement, according to described default
The error correction map table excavated in advance in environment carries out correction process to the raw tone read statement.
8. device according to claim 7, which is characterized in that it is described to recall processing module, it is specifically used for：
Speech recognition is carried out to the raw tone read statement, obtains corresponding original character sentence；
According to the original character sentence, attempt to carry out service from preset set of service to recall processing.
9. device according to claim 7, which is characterized in that the correction module is specifically used for according to the default ring
The error correction map table excavated in advance in border, the original character sentence corresponding to the raw tone read statement carry out
Correction process obtains the corresponding target text sentence of the raw tone read statement.
10. device according to claim 9, which is characterized in that it is described to recall processing module, it is additionally operable to according to the target
Word sentence, attempts to carry out service from the set of service to recall processing.
11. the device described in claim 7, which is characterized in that the receiving module is specifically used for receiving in the default environment
Intelligent terminal send the user raw tone read statement, the raw tone read statement of the user is institute
State intelligent terminal acquisition.
12. according to any devices of claim 7-11, which is characterized in that described device further includes：
Acquisition module, for acquiring in the default environment, in the preset acquisition time period, either preset times is all
Voice read statement and all voice read statements in can recall at least one correct voice of respective service
Read statement；
Module is excavated, for excavating the difference with each correct voice read statement from all voice read statements
Degree is less than time occurred jointly with the correct voice read statement in default diversity factor threshold value, and/or the collection period
Garbled voice read statement of the number more than preset times threshold value；
Module is established, for will be established between each correct voice read statement and the corresponding garbled voice read statement
Error correction map relationship obtains the error correction map table.
13. device according to claim 12, which is characterized in that error correcting handling arrangement setting in terminal device or
In cloud server.
14. a kind of computer equipment, which is characterized in that the equipment includes：
One or more processors；
Memory, for storing one or more programs；
When one or more of programs are executed by one or more of processors so that one or more of processors are real
The now method as described in any in claim 1-6.
15. a kind of computer-readable medium, is stored thereon with computer program, which is characterized in that the program is executed by processor
Methods of the Shi Shixian as described in any in claim 1-6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810225708.5A CN108595412B (en) | 2018-03-19 | 2018-03-19 | Error correction processing method and device, computer equipment and readable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810225708.5A CN108595412B (en) | 2018-03-19 | 2018-03-19 | Error correction processing method and device, computer equipment and readable medium |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108595412A true CN108595412A (en) | 2018-09-28 |
| CN108595412B CN108595412B (en) | 2020-03-27 |
Family
ID=63626615
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810225708.5A Active CN108595412B (en) | 2018-03-19 | 2018-03-19 | Error correction processing method and device, computer equipment and readable medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108595412B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109243433A (en) * | 2018-11-06 | 2019-01-18 | 北京百度网讯科技有限公司 | Audio recognition method and device |
| CN109686365A (en) * | 2018-12-26 | 2019-04-26 | 深圳供电局有限公司 | A kind of audio recognition method and speech recognition system |
| CN110415679A (en) * | 2019-07-25 | 2019-11-05 | 北京百度网讯科技有限公司 | Voice error correction method, device, equipment and storage medium |
| CN112541342A (en) * | 2020-12-08 | 2021-03-23 | 北京百度网讯科技有限公司 | Text error correction method and device, electronic equipment and storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1356628A (en) * | 2000-07-05 | 2002-07-03 | 国际商业机器公司 | Speech recognition correction for equipment wiht limited or no displays |
| CN102682763A (en) * | 2011-03-10 | 2012-09-19 | 北京三星通信技术研究有限公司 | Method, device and terminal for correcting named entity vocabularies in voice input text |
| CN102831177A (en) * | 2012-07-31 | 2012-12-19 | 聚熵信息技术（上海）有限公司 | Statement error correction method and system |
| CN107728783A (en) * | 2017-09-25 | 2018-02-23 | 联想(北京)有限公司 | Artificial intelligence process method and its system |
-
2018
- 2018-03-19 CN CN201810225708.5A patent/CN108595412B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1356628A (en) * | 2000-07-05 | 2002-07-03 | 国际商业机器公司 | Speech recognition correction for equipment wiht limited or no displays |
| CN102682763A (en) * | 2011-03-10 | 2012-09-19 | 北京三星通信技术研究有限公司 | Method, device and terminal for correcting named entity vocabularies in voice input text |
| CN102831177A (en) * | 2012-07-31 | 2012-12-19 | 聚熵信息技术（上海）有限公司 | Statement error correction method and system |
| CN107728783A (en) * | 2017-09-25 | 2018-02-23 | 联想(北京)有限公司 | Artificial intelligence process method and its system |
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109243433A (en) * | 2018-11-06 | 2019-01-18 | 北京百度网讯科技有限公司 | Audio recognition method and device |
| CN109243433B (en) * | 2018-11-06 | 2021-07-09 | 北京百度网讯科技有限公司 | Speech recognition method and device |
| CN109686365A (en) * | 2018-12-26 | 2019-04-26 | 深圳供电局有限公司 | A kind of audio recognition method and speech recognition system |
| CN109686365B (en) * | 2018-12-26 | 2021-07-13 | 深圳供电局有限公司 | Voice recognition method and voice recognition system |
| CN110415679A (en) * | 2019-07-25 | 2019-11-05 | 北京百度网讯科技有限公司 | Voice error correction method, device, equipment and storage medium |
| CN110415679B (en) * | 2019-07-25 | 2021-12-17 | 北京百度网讯科技有限公司 | Voice error correction method, device, equipment and storage medium |
| US11328708B2 (en) | 2019-07-25 | 2022-05-10 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Speech error-correction method, device and storage medium |
| CN112541342A (en) * | 2020-12-08 | 2021-03-23 | 北京百度网讯科技有限公司 | Text error correction method and device, electronic equipment and storage medium |
| CN112541342B (en) * | 2020-12-08 | 2022-07-22 | 北京百度网讯科技有限公司 | Text error correction method and device, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108595412B (en) | 2020-03-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10614803B2 (en) | Wake-on-voice method, terminal and storage medium | |
| CN106887225B (en) | Acoustic feature extraction method and device based on convolutional neural network and terminal equipment | |
| CN108595412A (en) | Correction processing method and device, computer equipment and readable medium | |
| CN108564966B (en) | Voice test method and device with storage function | |
| CN104050966B (en) | The voice interactive method of terminal device and the terminal device for using this method | |
| CN112349273A (en) | Speech synthesis method based on speaker, model training method and related equipment | |
| CN108133707A (en) | A kind of content share method and system | |
| CN108470034A (en) | A kind of smart machine service providing method and system | |
| CN110381221B (en) | Call processing method, device, system, equipment and computer storage medium | |
| CN108269567A (en) | For generating the method, apparatus of far field voice data, computing device and computer readable storage medium | |
| CN109545193A (en) | Method and apparatus for generating model | |
| CN104462058B (en) | Character string identification method and device | |
| CN109274831A (en) | A kind of audio communication method, device, equipment and readable storage medium storing program for executing | |
| CN101807399A (en) | Voice recognition method and device | |
| CN104866308A (en) | Scenario image generation method and apparatus | |
| CN107240396B (en) | Speaker self-adaptation method, device, equipment and storage medium | |
| CN108391020A (en) | A kind of call control method, device, equipment and storage medium | |
| WO2021227308A1 (en) | Video resource generation method and apparatus | |
| CN111462726B (en) | Method, device, equipment and medium for answering out call | |
| CN103839547A (en) | System for loading corresponding instruction elements by comparing voice operation signals and method thereof | |
| CN109545203A (en) | Audio recognition method, device, equipment and storage medium | |
| CN111507698A (en) | Processing method and device for transferring accounts, computing equipment and medium | |
| CN110365371A (en) | The method and its system, electronic equipment that trigger signal realizes translation system control are provided based on bluetooth equipment | |
| CN114093346A (en) | Joint automatic speech recognition and text-to-speech conversion using an antagonistic neural network | |
| CN108597499A (en) | Method of speech processing and voice processing apparatus |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20210508 Address after: 100085 Baidu Building, 10 Shangdi Tenth Street, Haidian District, Beijing Patentee after: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY Co.,Ltd. Patentee after: Shanghai Xiaodu Technology Co.,Ltd. Address before: 100085 Baidu Building, 10 Shangdi Tenth Street, Haidian District, Beijing Patentee before: BEIJING BAIDU NETCOM SCIENCE AND TECHNOLOGY Co.,Ltd. |