CN111833877A - Method and device for repairing voice recognition single number, electronic equipment and storage medium - Google Patents

Method and device for repairing voice recognition single number, electronic equipment and storage medium Download PDF

Info

Publication number
CN111833877A
CN111833877A CN202010692620.1A CN202010692620A CN111833877A CN 111833877 A CN111833877 A CN 111833877A CN 202010692620 A CN202010692620 A CN 202010692620A CN 111833877 A CN111833877 A CN 111833877A
Authority
CN
China
Prior art keywords
single number
repair
identification
character
current
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
Application number
CN202010692620.1A
Other languages
Chinese (zh)
Other versions
CN111833877B (en
Inventor
杨柳
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
AI Speech Ltd
Original Assignee
AI Speech Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by AI Speech Ltd filed Critical AI Speech Ltd
Priority to CN202010692620.1A priority Critical patent/CN111833877B/en
Publication of CN111833877A publication Critical patent/CN111833877A/en
Application granted granted Critical
Publication of CN111833877B publication Critical patent/CN111833877B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00Speech recognition
    • G10L15/22Procedures used during a speech recognition process, e.g. man-machine dialogue
    • GPHYSICS
    • G10MUSICAL INSTRUMENTS; ACOUSTICS
    • G10LSPEECH ANALYSIS OR SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00Speech recognition
    • G10L15/26Speech to text systems

Abstract

The invention discloses a method and a device for repairing a voice recognition single number, electronic equipment and a storage medium, wherein the method comprises the following steps: the method comprises the steps that terminal equipment obtains a voice signal and carries out voice recognition on the voice signal to obtain a recognition single number; the terminal equipment records the identification single number into a database, judges whether the identification single number is a first input single number or not, and if yes, executes first single number repair; if not, the terminal equipment judges whether to execute user repair or not, and executes automatic repair according to the judgment result or executes the user repair first and then executes the automatic repair; generating one or more repair order numbers; and the terminal equipment determines whether the correct single number exists in the one or more repair single numbers or not based on the verification result of the single number system. According to the method provided by the embodiment of the invention, an automatic repairing mode is set, so that the interaction times of the user with the application and the system can be reduced, the user experience is improved, and the accuracy of single number correction is improved.

Description

Method and device for repairing voice recognition single number, electronic equipment and storage medium
Technical Field
The present invention relates to the field of speech recognition technologies, and in particular, to a method and an apparatus for repairing a speech recognition single number, an electronic device, and a storage medium.
Background
With the continuous development of speech recognition technology, more and more application scenes of speech interaction and man-machine conversation appear. For example, many applications require a user to provide information similar to a single number, such as an order number, a license plate number, an identification number, a bank card number, etc., which can be provided by the user as a voice input in a voice interaction scenario in order to facilitate the user to provide the information.
More and more service scenes need to identify the single number information input by the user in a voice form, but the single number information is usually long letter and number string information, and because the length of the information is long, the user cannot complete and accurately speak at one time when inputting voice, and because the voice recognition may generate errors, the final recognition result may be inaccurate.
In order to solve the problem that the single number information is long and cannot accurately pass through voice recognition, the current solution is that after the single number information is recognized through a voice recognition technology, a user confirms the single number, and when a recognition error is found, the user actively modifies the single number information to correct the error information in the recognized single number information. However, this way of active modification by the user increases the number of interactions between the user and the application and system, reduces the user experience, and may introduce new errors due to the speech recognition errors that may be brought by the speech recognition technology.
Disclosure of Invention
The embodiment of the invention provides a method and a device for repairing a voice recognition single number, electronic equipment and a storage medium, which are used for solving at least one of the technical problems.
In a first aspect, an embodiment of the present invention provides a method for repairing a speech recognition single number, where the method is applied to a terminal device, and the method includes:
s11, the terminal equipment acquires a voice signal and performs voice recognition on the voice signal to obtain a recognition single number;
s12, the terminal device records the identification single number into a database, judges whether the identification single number is a first input single number or not, and if yes, enters S13; if not, the process goes to S14;
s13, the terminal equipment executes first single number repair to generate one or more repair single numbers, and the step enters S17;
s14, the terminal equipment judges whether to execute user repair, if so, the process goes to S15; if not, the process goes to S16;
s15, the terminal equipment executes user repair;
s16, the terminal equipment executes automatic repair to generate one or more repair single numbers, wherein the automatic repair comprises at least one of similarity repair, splicing repair, de-duplication repair and error-prone character repair;
s17, the terminal equipment determines whether a correct single number exists in the one or more repair single numbers based on the verification result of the single number system, and if so, the correct single number is recorded; if not, the process proceeds to S14.
In a second aspect, an embodiment of the present invention provides a device for repairing a speech recognition single number, which is applied to a terminal device, and includes:
the recognition module is configured to acquire a voice signal and perform voice recognition on the voice signal to obtain a recognition single number;
the first judgment module is configured to record the identification single number into a database and judge whether the identification single number is a first input single number;
a first repair module configured to perform a first single number repair to generate one or more repair single numbers;
a second judgment module configured to judge whether to execute user repair;
a user repair module configured to perform user repair;
the automatic repair module is configured to execute automatic repair and generate one or more repair single numbers, wherein the automatic repair comprises at least one of similarity repair, splicing repair, de-duplication repair and error-prone character repair;
and the verification module is configured to determine whether a correct single number exists in the one or more repair single numbers based on a verification result of the single number system.
In a third aspect, an electronic device is provided, comprising: the device comprises at least one processor and a memory which is in communication connection with the at least one processor, wherein the memory stores instructions which can be executed by the at least one processor, and the instructions are executed by the at least one processor so as to enable the at least one processor to execute any one of the voice recognition single number repairing methods of the invention.
In a fourth aspect, an embodiment of the present invention provides a storage medium, where one or more programs including execution instructions are stored, where the execution instructions can be read and executed by an electronic device (including but not limited to a computer, a server, or a network device, etc.) to perform any one of the above methods for repairing a speech recognition single number of the present invention.
The embodiment of the invention has the beneficial effects that: compared with the prior art, the method provided by the embodiment of the invention repairs the single number identified by the voice through the automatic repairing mode; the automatic repairing mode is based on the single number recorded in the database, similarity repairing is carried out, and splicing repairing, duplication removing repairing and error-prone character repairing are carried out. Because the automatic repairing mode is set, the interaction times of the user with the application and the system can be reduced, the user experience is improved, and the accuracy of single number correction is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1A is a flow chart of one embodiment of a method for repairing a speech recognition single number according to the present invention;
FIG. 1B is a flowchart of an embodiment of a method for repairing a speech recognition single number according to the present invention;
FIG. 2 is a schematic structural diagram illustrating an embodiment of a method for repairing a speech recognition single number according to the present invention;
fig. 3 is a schematic structural diagram of an embodiment of an electronic device according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict.
The invention may be described in the general context of computer-executable instructions, such as program modules, being executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices that are linked through a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media including memory storage devices.
As used in this disclosure, "module," "device," "system," and the like are intended to refer to a computer-related entity, either hardware, a combination of hardware and software, or software in execution. In particular, for example, an element may be, but is not limited to being, a process running on a processor, an object, an executable, a thread of execution, a program, and/or a computer. Also, an application or script running on a server, or a server, may be an element. One or more elements may be in a process and/or thread of execution and an element may be localized on one computer and/or distributed between two or more computers and may be operated by various computer-readable media. The elements may also communicate by way of local and/or remote processes based on a signal having one or more data packets, e.g., from a data packet interacting with another element in a local system, distributed system, and/or across a network in the internet with other systems by way of the signal.
Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
As shown in fig. 1A, an embodiment of the present invention provides a method for repairing a speech recognition single number, which is applied to a terminal device, and the method includes:
s11, the terminal equipment acquires a voice signal and performs voice recognition on the voice signal to obtain a recognition single number;
s12, the terminal device records the identification single number into a database, judges whether the identification single number is a first input single number or not, and if yes, enters S13; if not, the process goes to S14;
s13, the terminal equipment executes first single number repair to generate one or more repair single numbers, and the step enters S17;
s14, the terminal equipment judges whether to execute user repair, if so, the process goes to S15; if not, the process goes to S16;
s15, the terminal equipment executes user repair;
s16, the terminal equipment executes automatic repair to generate one or more repair single numbers, wherein the automatic repair comprises at least one of similarity repair, splicing repair, de-duplication repair and error-prone character repair;
s17, the terminal equipment determines whether a correct single number exists in the one or more repair single numbers based on the verification result of the single number system, and if so, the correct single number is recorded; if not, the process proceeds to S14.
The S11, the terminal device obtains a voice signal, and carries out voice recognition on the voice signal to obtain a recognition single number;
the terminal device can be any intelligent terminal or electronic device such as a smart phone, a tablet computer and a PC.
In the embodiment, the single number in the user voice signal is recognized through ASR.
In this embodiment, the length of the single sign in the user speech signal conforms to the normal distribution. For example, if the correct single number is 10 digits, the highest one of the identified single numbers is the 10 digit single number, and the second is the 9 and 11 digits, which are sequentially decreased. For another example, if the number of digits of the single number is correct, if the identified single number has an error, the ratio of digits with one bit error is the largest, and so on.
S12, the terminal device records the identification single number into a database, judges whether the identification single number is a first input single number or not, and if yes, enters S13; if not, the process proceeds to S14,
the terminal device records the identification single number into a database, and judges whether the identification single number is a first input single number, and the method comprises the following steps:
the terminal equipment records the identification single number and the user voice instruction identification ID into the database, inquires the database according to the user voice instruction identification ID, determines whether the identification single number has the single number, and determines that the identification single number is the first input single number if the identification single number does not have the single number; otherwise, the identification single number is not the first input single number; the above single number is a unique single number obtained by searching through the user voice instruction identification ID.
S13, the terminal equipment executes first single number repair to generate one or more repair single numbers, and the step enters S17;
wherein, the first single number repair comprises duplication removal repair and error-prone character repair;
the duplicate removal and repair is that if the number of the character digits of the identification single number and/or the repair single number exceeds the correct number, the terminal equipment searches for the repeated sub-character strings in the identification single number and/or the repair single number, and if the number of the character strings obtained by deleting the repeated sub-character strings in the identification single number and/or the repair single number is the correct number, the repeated sub-character strings in the identification single number and/or the repair single number are deleted, so that one or more repair single numbers are obtained.
Deduplication fixes addresses the situation where the user has duplicate readings while saying a single number. For example, the correct single number to be input by the user is "1234567890", the voice input by the user is "single number is 123, 1234567890", it is determined that the number digit number of the single number exceeds the correct number of digits by 10 for the recognized single number "1231234567890", and the repeated sub-number string "123" exists in the single number, and the repeated sub-number string "123" is deleted, so as to obtain the repaired single number "1234567890".
And the repeated repairing also comprises deleting the corresponding repeated sub-digit strings if the repeated sub-digit strings exist at a plurality of positions and the digit number of the obtained digit string is the correct digit number after the repeated sub-digit strings are respectively deleted.
The error-prone character repairing is to store the identification single number and/or the repair single number by the terminal equipment, count and determine characters which are easy to confuse and lose according to historical data, and replace the characters which are easy to confuse in the identification single number and/or the repair single number; supplementing characters which are easy to lose to the identification single number and/or the repair single number with insufficient digits; and generating a new repair order number, and taking the new repair order number and the pre-stored identification order number and/or repair order number as target repair order numbers so as to obtain one or more repair order numbers.
And if the identification single number is not changed after the duplication removal repair and the error-prone character repair, taking the identification single number as a repair single number.
In another embodiment, for example, according to the historical data, it is counted that the ratio of confusion of 3 and 4 exceeds the threshold, and it is counted that the ratio of missing numbers 2 and 5 exceeds the threshold, then 3 and 4 are determined to be confusable numbers; 2 and 5 are determined as easy-missing numbers. And replacing the confusable number 3 in the single number and/or the repair single number with the number 4, replacing the number 4 with the number 3, and identifying all the confusable numbers in the single number and/or the repair single number by replacing part or all the confusable numbers in the single number and/or the repair single number by replacing the operation.
The S14, the terminal equipment judges whether to execute user repair, if yes, the S15 is entered; if not, the process proceeds to S16,
the terminal equipment receives user operation, judges whether to execute user repair according to the user operation, and executes the user repair if the user repair needs to be executed; otherwise, the terminal equipment executes automatic repair.
The S15, the terminal device performs user repair,
the user repairing is that the terminal equipment receives an active modification instruction of a user and modifies the active modification instruction in a local modification or full modification mode, wherein the local modification is that the expression of the user is received to modify a single character position until all characters which the user wants to modify are modified, and the modified single number is used as a repairing single number; and the total modification is to receive the voice signal of the whole single number re-input by the user, recognize the voice signal to obtain a new identification single number, and reset and cover the stored identification single number in the database by using the new identification single number as a repair single number.
And only one repair single number is generated in the local modification mode and the whole modification mode.
S16, the terminal device executes automatic repair to generate one or more repair single numbers, the automatic repair includes at least one of similarity repair, splicing repair, de-duplication repair and error-prone character repair,
wherein, the similarity refers to the number of the same numbers at the same positions of two single numbers;
the similarity restoration includes:
determining the correct number n of the single number, selecting the identification single number and/or the repair single number with the character number between [ n-1, n +1] as a single number set, and selecting a single number from the single number set as the current single number; the repairing objects of the similarity repairing of the terminal equipment are the current single number and the previous single number of the single number, and one or more repairing single numbers are generated according to a repairing strategy; until all the single numbers in the single number set and the corresponding single numbers above are repaired according to a repair strategy, wherein n is more than 1; the repair strategy comprises:
a first repair strategy is as follows: if the current sheet number and the previous sheet number of the sheet number are both n bits,
s101, comparing characters of each character position in the current single number and the previous single number of the single number, and determining all character position sets with different characters;
s102, randomly selecting a character bit with different characters from all the character bit sets, replacing the character of the corresponding character bit in the current single number with the character of the character bit corresponding to the single number above the single number, generating and recording a process repairing single number, and deleting the character bit from all the character bit sets;
s103, if the whole character bit set is not empty, entering S102; otherwise, entering S104;
s104, obtaining a repair order number;
and (2) repairing strategy two: if the current single number is n bits, the previous single number of the current single number is n +1 bits,
s201, marking the above single number of the current single number from left to right as 1 to n +1 bits, and setting i to be 1;
s202, deleting the ith bit of the previous single number of the current single number, generating a process repair single number, and calculating the similarity between the process repair single number and the current single number;
s203, if i is less than n +1, setting i as i +1, and entering S202; otherwise, entering S204;
s204, recording the process repair single number with the maximum similarity as the single number with the maximum similarity;
s205, repeating the first repair strategy for the single number with the maximum similarity and the current single number, wherein when the first repair strategy is repeated, the single number with the maximum similarity serves as the single number above, and the first repair strategy is executed with the current single number;
and (3) repairing strategy III: if the current single number is n-1 bits, the previous single number of the current single number is n bits,
s301, adding a placeholder at a certain position of the current single number to obtain a process repair single number, and calculating the similarity of the process repair single number and the previous single number of the current single number;
s302, repeating the step S301 until the obtained process with the maximum similarity restores the single number, wherein the position of the placeholder is the position of the missing character position;
and S303, replacing the corresponding placeholder with the character corresponding to the position of the missing character position and the previous single number of the current single number to obtain the repair single number.
In this embodiment, the similarity repair also supports repairing more cases and supports other repair strategies. For example, the above single numbers are n-1, n, and n +1 bits respectively, and the current single numbers are n-1, n, and n +1 bits respectively, which can generate 9 combinations, and the similarity repairing method is similar and will not be described herein again.
The following examples illustrate the repair of similarity repair:
assuming the correct single sign is a 10-digit number: 1234567890. next, several examples will be given to explain the similarity restoration method in detail.
Example 1:
the above sheet number: 1233567890
Current single number: 1234567891
The single number is 1233567890 above, where bit 4, misidentifies 4 as 3, the single number is misidentified, thus requiring the user to re-provide the single number. The current sheet number is obtained as 1234567891, and the last bit identifies an error. Since the current number is 10, it is likely to be the correct number, and therefore a list of repair numbers is added. And calculating the similarity of the above single number and the current single number to be 8, which indicates that the two single numbers are possibly staggered by one bit respectively, and if the two staggered positions are different, the terminal equipment can spell the correct single number. Therefore, the positions of the two single numbers which are different are found, the number of the current single number is replaced by the number of the last single number, and then two new single numbers are obtained: 1233567891, 1234567890. Thus, the list of final repair sheet numbers is: 1234567891, 1233567891, 1234567890. It can be seen that although the single number spoken by the user has errors in both times, the correct result is included in the list of repaired single numbers.
Example 2:
the above sheet number: 12345678990
Current single number: 1244567890
The above single sign is 12345678990 with one more digit, 9 more than the correct single sign. The current single number is 1244567890, misidentifying 3 as 4 in bit 3 as compared to the correct single number. As above, the current single number is 10 in length and therefore is likely to be the correct single number, which the terminal device adds to the list of repair single numbers. Since the above single sign is 11 bits, there is a high probability that there is one more bit. Deleting one digit of the document number respectively, and then calculating the similarity together with the second round of document numbers, it can be found that when deleting 9, the similarity is the highest, and the document number after the deleting operation is 1234567890. Then comparing the current single number with the repair single number and replacing different numbers, the single number 1234567890 can be obtained. The list of repair order numbers may be restored with 1244567890, 1234567890, including the correct result.
Example 3:
the above sheet number: 1233567890
Current single number: 123456789
The current single sign is 9 bits, and one bit is less, so the terminal equipment is respectively filled with the number at different positions to complement 10 bits, and then the similarity is calculated. And (4) replacing the filling position with the highest similarity, namely the correct filling position, with the number corresponding to the position of the above single sign to obtain a repair single sign 1234567890 which is also the correct single sign.
And splicing and repairing, if the character digit of the identification single number and/or the repair single number is lower than the correct digit, and the sum of the character digit of the identification single number and/or the repair single number and the digit of the above single number is the correct digit, splicing the identification single number and/or the repair single number to the above single number to obtain the repair single number.
For example, when a user inputs speech, a pause or a wrong interruption may occur, for example, if 4 bits are input for the first time and 6 bits are input for the second time, the two results may be spliced to obtain a repair single number.
The deduplication restoring and error-prone character restoring are the same as those in S13, and are not described herein again.
In this embodiment, the user repair is active repair of the user, so that the user performs partial modification or complete modification on the single number. The automatic repair is to obtain one or more repair single numbers according to the identified identification single number and/or the repair single number and the above information of the identification single number by a repair method such as similarity repair, and the correct single number may exist in the repair single numbers. Because the user repair can also introduce new identification errors, the automatic repair is executed after the user repair, and the accuracy of the single number repair can be improved. Therefore, the interaction times with the user are reduced, and the user experience is improved.
S17, the terminal equipment determines whether a correct single number exists in the one or more repair single numbers based on the verification result of the single number system, and if so, the correct single number is recorded; if not, go to S14;
the single number system can verify the repaired single number, judge whether the single number is correct or not, and obtain the correct single number through the verification of the single number system. The single number system can be a third-party system, for example, when the single number is an express single number, the single number system is an express single number system of an express company.
As yet another example, the repair of the speech recognition ticket number may be performed in the manner shown in FIG. 1B.
As shown in fig. 2, a speech recognition single number repairing apparatus according to another embodiment of the present invention is applied to a terminal device, and the apparatus includes:
the recognition module is configured to acquire a voice signal and perform voice recognition on the voice signal to obtain a recognition single number;
the first judgment module is configured to record the identification single number into a database and judge whether the identification single number is a first input single number;
a first repair module configured to perform a first single number repair to generate one or more repair single numbers;
a second judgment module configured to judge whether to execute user repair;
a user repair module configured to perform user repair;
the automatic repair module is configured to execute automatic repair and generate one or more repair single numbers, wherein the automatic repair comprises at least one of similarity repair, splicing repair, de-duplication repair and error-prone character repair;
and the verification module is configured to determine whether a correct single number exists in the one or more repair single numbers based on a verification result of the single number system.
It should be noted that for simplicity of explanation, the foregoing method embodiments are described as a series of acts or combination of acts, but those skilled in the art will appreciate that the present invention is not limited by the order of acts, as some steps may occur in other orders or concurrently in accordance with the invention. Further, those skilled in the art should also appreciate that the embodiments described in the specification are preferred embodiments and that the acts and modules referred to are not necessarily required by the invention.
In the foregoing embodiments, the descriptions of the respective embodiments have respective emphasis, and for parts that are not described in detail in a certain embodiment, reference may be made to related descriptions of other embodiments.
In some embodiments, the present invention further provides a storage medium having a computer program stored thereon, wherein the computer program is capable of executing the steps of the method for repairing a speech recognition ticket number when the computer program is executed by a processor.
In some embodiments, an embodiment of the present invention further provides an electronic device, which includes: the device comprises at least one processor and a memory which is in communication connection with the at least one processor, wherein the memory stores instructions which can be executed by the at least one processor, and the instructions are executed by the at least one processor so as to enable the at least one processor to execute a method for repairing a voice recognition single number.
The device for repairing the speech recognition single number according to the embodiment of the present invention can be used for executing the method for repairing the speech recognition single number according to the embodiment of the present invention, and accordingly achieves the technical effect achieved by the method for repairing the speech recognition single number according to the embodiment of the present invention, and further description is omitted here. In the embodiment of the present invention, the relevant functional module may be implemented by a hardware processor (hardware processor).
Fig. 3 is a schematic hardware structure diagram of an electronic device for a method for repairing a speech recognition single number according to another embodiment of the present application, and as shown in fig. 3, the device includes:
one or more processors 310 and a memory 320, one processor 310 being illustrated in fig. 3.
The apparatus for performing the repair method of the voice recognition ticket may further include: an input device 330 and an output device 340.
The processor 310, the memory 320, the input device 330, and the output device 340 may be connected by a bus or other means, such as the bus connection in fig. 3.
The memory 320 is a non-volatile computer-readable storage medium and can be used for storing non-volatile software programs, non-volatile computer-executable programs, and modules, such as program instructions/modules corresponding to the voice service method in the embodiment of the present application. The processor 310 executes various functional applications of the server and data processing by executing nonvolatile software programs, instructions and modules stored in the memory 320, namely, implements the voice service method of the above-described method embodiment.
The memory 320 may include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application program required for at least one function; the storage data area may store data created according to the use of the voice service apparatus, and the like. Further, the memory 320 may include high speed random access memory, and may also include non-volatile memory, such as at least one magnetic disk storage device, flash memory device, or other non-volatile solid state storage device. In some embodiments, memory 320 may optionally include memory located remotely from processor 310, which may be connected to a voice service device over a network. Examples of such networks include, but are not limited to, the internet, intranets, local area networks, mobile communication networks, and combinations thereof.
The input device 330 may receive input numeric or character information and generate signals related to user settings and function control of the voice service device. The output device 340 may include a display device such as a display screen.
The one or more modules are stored in the memory 320 and, when executed by the one or more processors 310, perform the voice service method of any of the method embodiments described above.
The product can execute the method provided by the embodiment of the application, and has the corresponding functional modules and beneficial effects of the execution method. For technical details that are not described in detail in this embodiment, reference may be made to the methods provided in the embodiments of the present application.
The electronic device of the embodiments of the present application exists in various forms, including but not limited to:
(1) mobile communication devices, which are characterized by mobile communication capabilities and are primarily targeted at providing voice and data communications. Such terminals include smart phones (e.g., iphones), multimedia phones, functional phones, and low-end phones, among others.
(2) The ultra-mobile personal computer equipment belongs to the category of personal computers, has calculation and processing functions and generally has the characteristic of mobile internet access. Such terminals include PDA, MID, and UMPC devices, such as ipads.
(3) Portable entertainment devices such devices may display and play multimedia content. Such devices include audio and video players (e.g., ipods), handheld game consoles, electronic books, as well as smart toys and portable car navigation devices.
(4) The server is similar to a general computer architecture, but has higher requirements on processing capability, stability, reliability, safety, expandability, manageability and the like because of the need of providing highly reliable services.
(5) And other electronic devices with data interaction functions.
The above-described embodiments of the apparatus are merely illustrative, and the units described as separate parts may or may not be physically separate, and parts displayed as units may or may not be physical units, may be located in one place, or may be distributed on a plurality of network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of the present embodiment.
Through the above description of the embodiments, those skilled in the art will clearly understand that each embodiment can be implemented by software plus a general hardware platform, and certainly can also be implemented by hardware. Based on such understanding, the technical solutions mentioned above may be embodied in the form of a software product, which may be stored in a computer-readable storage medium, such as ROM/RAM, magnetic disk, optical disk, etc., and includes a plurality of instructions for enabling a computer device (which may be a personal computer, a server, or a network device) to execute the method according to the embodiments or some parts of the embodiments.
Finally, it should be noted that: the above embodiments are only used to illustrate the technical solutions of the present application, and not to limit the same; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions in the embodiments of the present application.

Claims (14)

1. A method for repairing a voice recognition single number is applied to terminal equipment, and comprises the following steps:
s11, the terminal equipment acquires a voice signal and performs voice recognition on the voice signal to obtain a recognition single number;
s12, the terminal device records the identification single number into a database, judges whether the identification single number is a first input single number or not, and if yes, enters S13; if not, the process goes to S14;
s13, the terminal equipment executes first single number repair to generate one or more repair single numbers, and the step enters S17;
s14, the terminal equipment judges whether to execute user repair, if so, the process goes to S15; if not, the process goes to S16;
s15, the terminal equipment executes user repair;
s16, the terminal equipment executes automatic repair to generate one or more repair single numbers, wherein the automatic repair comprises at least one of similarity repair, splicing repair, de-duplication repair and error-prone character repair;
s17, the terminal equipment determines whether a correct single number exists in the one or more repair single numbers based on the verification result of the single number system, and if so, the correct single number is recorded; if not, the process proceeds to S14.
2. The method of claim 1, wherein the terminal device records the identification ticket number into a database, and determining whether the identification ticket number is a first-time input ticket number comprises:
the terminal equipment records the identification single number and the user voice instruction identification ID into the database, inquires the database according to the user voice instruction identification ID, determines whether the identification single number has the single number, and determines that the identification single number is the first input single number if the identification single number does not have the single number; otherwise, the identification single number is not the first input single number; the above single number is a unique single number obtained by searching through the user voice instruction identification ID.
3. The method of claim 1, wherein the first single number repair comprises a deduplication repair, an error prone character repair;
the de-duplication repair comprises the following steps: if the number of the character digits of the identification single number and/or the repair single number exceeds the correct number, the terminal equipment searches repeated sub character strings in the identification single number and/or the repair single number, and if the number of the character strings obtained by deleting the repeated sub character strings in the identification single number and/or the repair single number is the correct number, the repeated sub character strings in the identification single number and/or the repair single number are deleted, so that one or more repair single numbers are obtained;
the error-prone character is repaired as follows: for the identification single number and/or the repair single number, the terminal equipment stores the identification single number and/or the repair single number, counts and determines characters which are easy to confuse and lose according to historical data, and replaces the characters which are easy to confuse in the identification single number and/or the repair single number; supplementing characters which are easy to lose to the identification single number and/or the repair single number with insufficient digits; and generating a new repair order number, and taking the new repair order number and the pre-stored identification order number and/or repair order number as target repair order numbers so as to obtain one or more repair order numbers.
4. The method according to claim 1, wherein the user repair receives an active modification instruction of a user for the terminal device, and modifies the active modification instruction by adopting a local modification or full modification mode, wherein the local modification is to receive an expression of the user to modify a single character bit until all the characters which the user wants to modify are modified, and the modified single number is used as a repair single number; and the total modification is to receive the voice signal of the whole single number re-input by the user, recognize the voice signal to obtain a new identification single number, and reset and cover the stored identification single number in the database by using the new identification single number as a repair single number.
5. The method of claim 1, wherein similarity refers to the same number of characters in the same position of two single signs; the similarity restoration includes:
determining the correct number n of the single number, selecting the identification single number and/or the repair single number with the character number between [ n-1, n +1] as a single number set, and selecting a single number from the single number set as the current single number; the repair objects of the similarity repair are the current single number and the previous single number of the single number, and one or more repair single numbers are generated according to a repair strategy; until all the single numbers in the single number set and the corresponding single numbers above are repaired according to a repair strategy, wherein n is more than 1; the repair strategy comprises:
a first repair strategy is as follows: if the current sheet number and the previous sheet number of the sheet number are both n bits,
s101, comparing characters of each character position in the current single number and the previous single number of the single number, and determining all character position sets with different characters;
s102, randomly selecting a character bit with different characters from all the character bit sets, replacing the character of the corresponding character bit in the current single number with the character of the character bit corresponding to the single number above the single number, generating and recording a process repairing single number, and deleting the character bit from all the character bit sets;
s103, if the whole character bit set is not empty, entering S102; otherwise, entering S104;
s104, obtaining a repair order number;
and (2) repairing strategy two: if the current single number is n bits, the previous single number of the current single number is n +1 bits,
s201, marking the above single number of the current single number from left to right as 1 to n +1 bits, and setting i to be 1;
s202, deleting the ith bit of the previous single number of the current single number, generating a process repair single number, and calculating the similarity between the process repair single number and the current single number;
s203, if i is less than n +1, setting i as i +1, and entering S202; otherwise, entering S204;
s204, recording the process repair single number with the maximum similarity as the single number with the maximum similarity;
s205, repeating the first repair strategy for the single number with the maximum similarity and the current single number, wherein when the first repair strategy is repeated, the single number with the maximum similarity serves as the single number above, and the first repair strategy is executed with the current single number;
and (3) repairing strategy III: if the current single number is n-1 bits, the previous single number of the current single number is n bits,
s301, adding a placeholder at a certain position of the current single number to obtain a process repair single number, and calculating the similarity of the process repair single number and the previous single number of the current single number;
s302, repeating the step S301 until the obtained process with the maximum similarity restores the single number, wherein the position of the placeholder is the position of the missing character position;
and S303, replacing the corresponding placeholder with the character corresponding to the position of the missing character position and the previous single number of the current single number to obtain the repair single number.
6. The method of claim 1, wherein the splice repair is: and if the character number of the identification single number and/or the repair single number is lower than the correct number, and the sum of the character number of the identification single number and/or the repair single number and the number of the single number above is the correct number, splicing the identification single number and/or the repair single number to the single number above to obtain the repair single number.
7. A speech recognition single number repairing device is applied to terminal equipment and comprises:
the recognition module is configured to acquire a voice signal and perform voice recognition on the voice signal to obtain a recognition single number;
the first judgment module is configured to record the identification single number into a database and judge whether the identification single number is a first input single number;
a first repair module configured to perform a first single number repair to generate one or more repair single numbers;
a second judgment module configured to judge whether to execute user repair;
a user repair module configured to perform user repair;
the automatic repair module is configured to execute automatic repair and generate one or more repair single numbers, wherein the automatic repair comprises at least one of similarity repair, splicing repair, de-duplication repair and error-prone character repair;
and the verification module is configured to determine whether a correct single number exists in the one or more repair single numbers based on a verification result of the single number system.
8. The apparatus of claim 7, wherein the first determining module comprises:
the terminal equipment records the identification single number and the user voice instruction identification ID into the database, inquires the database according to the user voice instruction identification ID, determines whether the identification single number has the single number, and determines that the identification single number is the first input single number if the identification single number does not have the single number; otherwise, the identification single number is not the first input single number; the above single number is a unique single number obtained by searching through the user voice instruction identification ID.
9. The apparatus of claim 7, wherein the first single number repair comprises a deduplication repair, an error prone character repair;
the de-duplication repair comprises the following steps: if the number of the character digits of the identification single number and/or the repair single number exceeds the correct number, the terminal equipment searches repeated sub character strings in the identification single number and/or the repair single number, and if the number of the character strings obtained by deleting the repeated sub character strings in the identification single number and/or the repair single number is the correct number, the repeated sub character strings in the identification single number and/or the repair single number are deleted, so that one or more repair single numbers are obtained;
the error-prone character is repaired as follows: for the identification single number and/or the repair single number, the terminal equipment stores the identification single number and/or the repair single number, counts and determines characters which are easy to confuse and lose according to historical data, and replaces the characters which are easy to confuse in the identification single number and/or the repair single number; supplementing characters which are easy to lose to the identification single number and/or the repair single number with insufficient digits; and generating a new repair order number, and taking the new repair order number and the pre-stored identification order number and/or repair order number as target repair order numbers so as to obtain one or more repair order numbers.
10. The device according to claim 7, wherein the user repair receives an active modification instruction of a user for the terminal device, and modifies the active modification instruction by adopting a local modification or full modification mode, wherein the local modification is to receive an expression of the user to modify a single character bit until all the characters which the user wants to modify are modified, and takes the modified single number as a repair single number; and the total modification is to receive the voice signal of the whole single number re-input by the user, recognize the voice signal to obtain a new identification single number, and reset and cover the stored identification single number in the database by using the new identification single number as a repair single number.
11. The apparatus of claim 10, wherein the automated repair module comprises:
the first determining submodule is configured to determine the correct digit n of the single number, select the identification single number and/or the repair single number with the character digit between [ n-1, n +1] as a single number set, and select a single number from the single number set as the current single number; the repair objects of the similarity repair are the current single number and the previous single number of the single number, and one or more repair single numbers are generated according to a repair strategy; until all the single numbers in the single number set and the corresponding single numbers above are repaired according to a repair strategy, wherein n is more than 1;
a first repair submodule, comprising:
a first comparison submodule: if the current single number and the previous single number of the single number are both n bits, comparing the characters of each character bit in the current single number and the previous single number of the single number, and determining all character bit sets with different characters;
a first alternative sub-module: randomly selecting a character bit with different characters from all the character bit sets, replacing the character of the corresponding character bit in the current single number with the character of the character bit corresponding to the previous single number of the single number, generating and recording a process repairing single number, and deleting the character bit from all the character bit sets;
a first judgment sub-module: judging whether all the character bit sets are empty or not;
the first repair single number storage submodule: obtaining a repair order number;
a second repair submodule, comprising:
a second determination sub-module: if the current single number is n bits, the previous single number of the current single number is n +1 bits, the previous single number of the current single number is marked as 1 to n +1 bits from left to right, and i is set to be 1;
a second deletion submodule: deleting the ith bit of the previous single number of the current single number, generating a process repair single number, and calculating the similarity between the process repair single number and the current single number;
a second judgment sub-module: if i < n +1, setting i as i + 1;
the second single number storage submodule: recording the process repair single number with the maximum similarity as the single number with the maximum similarity;
the second repair calculation submodule repeats the first repair strategy for the single number with the maximum similarity and the current single number, and when the first repair strategy is repeated, the single number with the maximum similarity serves as the single number above and executes the first repair strategy together with the current single number;
a third repair submodule, comprising:
a third judgment submodule, wherein if the current single number is n-1 bits, the previous single number of the current single number is n bits;
a third adding sub-module, adding a placeholder at a certain position of the current single number to obtain a process repairing single number, and calculating the similarity of the process repairing single number and the previous single number of the current single number;
repeating the third increasing sub-module until the obtained process with the maximum similarity restores the single number, wherein the position of the placeholder is the position of the missing character bit;
the third repair single number storage submodule: and replacing the corresponding placeholder by using the character corresponding to the position of the above single number of the current single number and the missing character position to obtain a repair single number.
12. The apparatus of claim 7, wherein the splice repair is: and if the character number of the identification single number and/or the repair single number is lower than the correct number, and the sum of the character number of the identification single number and/or the repair single number and the number of the single number above is the correct number, splicing the identification single number and/or the repair single number to the single number above to obtain the repair single number.
13. An electronic device, comprising: at least one processor, and a memory communicatively coupled to the at least one processor, wherein the memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps of the method of any one of claims 1-6.
14. A storage medium on which a computer program is stored which, when being executed by a processor, carries out the steps of the method of any one of claims 1 to 6.
CN202010692620.1A 2020-07-17 2020-07-17 Method and device for repairing voice recognition single number, electronic equipment and storage medium Active CN111833877B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010692620.1A CN111833877B (en) 2020-07-17 2020-07-17 Method and device for repairing voice recognition single number, electronic equipment and storage medium

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010692620.1A CN111833877B (en) 2020-07-17 2020-07-17 Method and device for repairing voice recognition single number, electronic equipment and storage medium

Publications (2)

Publication Number Publication Date
CN111833877A true CN111833877A (en) 2020-10-27
CN111833877B CN111833877B (en) 2022-03-29

Family

ID=72923473

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010692620.1A Active CN111833877B (en) 2020-07-17 2020-07-17 Method and device for repairing voice recognition single number, electronic equipment and storage medium

Country Status (1)

Country Link
CN (1) CN111833877B (en)

Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030125945A1 (en) * 2001-12-14 2003-07-03 Sean Doyle Automatically improving a voice recognition system
CN1875400A (en) * 2003-11-07 2006-12-06 佳能株式会社 Information processing apparatus, information processing method and recording medium, and program
US20120303368A1 (en) * 2011-05-27 2012-11-29 Ting Ma Number-assistant voice input system, number-assistant voice input method for voice input system and number-assistant voice correcting method for voice input system
CN105206260A (en) * 2015-08-31 2015-12-30 努比亚技术有限公司 Terminal voice broadcasting method, device and terminal voice operation method
CN107239685A (en) * 2017-06-02 2017-10-10 深圳市冠旭电子股份有限公司 Control method, bluetooth equipment and the computer-readable recording medium of bluetooth equipment
CN107622054A (en) * 2017-09-26 2018-01-23 科大讯飞股份有限公司 The error correction method and device of text data
CN107731229A (en) * 2017-09-29 2018-02-23 百度在线网络技术(北京)有限公司 Method and apparatus for identifying voice

Patent Citations (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20030125945A1 (en) * 2001-12-14 2003-07-03 Sean Doyle Automatically improving a voice recognition system
CN1875400A (en) * 2003-11-07 2006-12-06 佳能株式会社 Information processing apparatus, information processing method and recording medium, and program
US20120303368A1 (en) * 2011-05-27 2012-11-29 Ting Ma Number-assistant voice input system, number-assistant voice input method for voice input system and number-assistant voice correcting method for voice input system
CN105206260A (en) * 2015-08-31 2015-12-30 努比亚技术有限公司 Terminal voice broadcasting method, device and terminal voice operation method
CN107239685A (en) * 2017-06-02 2017-10-10 深圳市冠旭电子股份有限公司 Control method, bluetooth equipment and the computer-readable recording medium of bluetooth equipment
CN107622054A (en) * 2017-09-26 2018-01-23 科大讯飞股份有限公司 The error correction method and device of text data
CN107731229A (en) * 2017-09-29 2018-02-23 百度在线网络技术(北京)有限公司 Method and apparatus for identifying voice

Also Published As

Publication number Publication date
CN111833877B (en) 2022-03-29

Similar Documents

Publication Publication Date Title
CN108549662B (en) Complementary digestion method and device for semantic analysis results in multi-round conversation
CN106485261B (en) Image recognition method and device
CN104346372A (en) Method and device for evaluating prediction models
CN111339758B (en) Text error correction method and system based on deep learning model
CN110781960B (en) Training method, classification method, device and equipment of video classification model
CN109634912B (en) Data migration method and system
CN111833877B (en) Method and device for repairing voice recognition single number, electronic equipment and storage medium
CN110837635A (en) Method, device, equipment and storage medium for equipment verification
CN109273004B (en) Predictive speech recognition method and device based on big data
CN109299231B (en) Dialog state tracking method, system, electronic device and storage medium
CN111159169A (en) Data management method and equipment
CN114722245A (en) Application question correcting method and device, storage medium and electronic equipment
CN111858966B (en) Knowledge graph updating method and device, terminal equipment and readable storage medium
CN112686732B (en) Abnormal address data identification method, device, equipment and medium
CN114595318A (en) Customer service reply quality evaluation method and system
CN114065762A (en) Text information processing method, device, medium and equipment
CN109658940B (en) Method and system for updating voice recognition resources
CN115826992A (en) Method for identifying abnormal APK, electronic device and readable storage medium
CN105278993A (en) Linux system based drive module upgrading method and apparatus
CN114218937B (en) Data error correction method and device and electronic equipment
CN113609352B (en) Character string retrieval method, device, computer equipment and storage medium
CN110597517A (en) Method and device for dynamically analyzing mat file
CN111199156B (en) Named entity recognition method, device, storage medium and processor
US10467392B2 (en) Verifying and controlling computer system state using pronounceable recognizable hash words
US20220319505A1 (en) System and method for rapid improvement of virtual speech agent&#39;s natural language understanding

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
CB02 Change of applicant information

Address after: 215123 building 14, Tengfei Innovation Park, 388 Xinping street, Suzhou Industrial Park, Suzhou City, Jiangsu Province

Applicant after: Sipic Technology Co.,Ltd.

Address before: 215123 building 14, Tengfei Innovation Park, 388 Xinping street, Suzhou Industrial Park, Suzhou City, Jiangsu Province

Applicant before: AI SPEECH Ltd.

CB02 Change of applicant information
GR01 Patent grant
GR01 Patent grant