CN116129904A - Speech recognition method, device, electronic equipment and storage medium - Google Patents

Abstract

The application discloses a voice recognition method, device, electronic equipment and storage medium, belonging to the technical field of voice recognition. The speech recognition method includes: in response to the received speech data, converting the speech data into first text data; obtaining the pronunciation information of the first phrase data in the first text data; according to the pronunciation information and the first corresponding relationship, for the first The first phrase data in the text data is added with an identification code to obtain the second text data, wherein the first correspondence is the correspondence between the pronunciation information and the identification code, and each identification code corresponds to at least two pronunciation information; based on the second text data Identify and identify the corresponding control instructions of the voice data.

Description

Speech recognition method, device, electronic device and storage medium

technical field

The present application belongs to the technical field of speech recognition, and in particular relates to a speech recognition method, device, electronic equipment and storage medium.

Background technique

Human-machine dialogue is an important field in natural language processing, such as robot customer service, ordering food, booking tickets, etc. The most important thing is to let the machine understand the meaning of human speech, that is, natural language understanding.

The voice recognition function in the related art has the problem of insufficient generalization, and the voice command cannot be accurately recognized when the pronunciation of the received voice command is deviated.

Contents of the invention

The purpose of the embodiments of the present application is to provide a voice recognition method, device, electronic equipment and storage medium, which improves the accuracy of the electronic equipment in recognizing voice commands that are not pronounced correctly.

In the first aspect, the embodiment of the present application provides a voice recognition method, including: receiving voice data, converting the voice data into first text data; acquiring the pronunciation information of the first phrase data in the first text data; according to the pronunciation information and The first correspondence is to add an identification code to the first phrase data in the first text data to obtain the second text data, wherein the first correspondence is the correspondence between the pronunciation information and the identification code, and each identification code corresponds to at least two Pronunciation information: based on the second text data recognition, identify the control instruction corresponding to the voice data.

In a second aspect, the embodiment of the present application provides a voice recognition device, including: a processing module, configured to receive voice data, and convert the voice data into first text data; an acquisition module, configured to acquire the first text data in the first text data; The pronunciation information of phrase data; Processing module, for according to pronunciation information and the first correspondence, add identification code to the first phrase data in the first text data, obtain the second text data, wherein, the first correspondence is pronunciation information Corresponding relationship with the identification codes, each identification code corresponds to at least two reading information; the processing module is used to recognize the corresponding control instructions of the voice data based on the second text data recognition.

In the third aspect, the embodiment of the present application provides an electronic device, including: a processor and a memory, the memory stores programs or instructions that can be run on the processor, and when the programs or instructions are executed by the processor, the speech as described in the first aspect is realized. Identify the steps of the method.

In a fourth aspect, the embodiment of the present application provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by a processor, the steps of the speech recognition method in the first aspect are implemented.

In the fifth aspect, the embodiment of the present application provides a chip, the chip includes a processor and a communication interface, the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the speech recognition method as in the first aspect A step of.

In a sixth aspect, an embodiment of the present application provides a computer program product, the program product is stored in a storage medium, and the program product is executed by at least one processor to implement the steps of the speech recognition method in the first aspect.

In the embodiment of the present application, when the electronic device receives the voice data, it converts the received voice data into the first text data, and adds an identification matching the pronunciation information to the first phrase data in the first text data encoding to generate the second text data, because the identification code in the second text data corresponds to a plurality of similar pronunciation information, so that the electronic device can determine the true meaning of the first phrase data based on the identification code, realizing that even if the electronic device receives Even when the pronunciation of the first phrase data is inaccurate, the true intention of the voice data including the first phrase data can still be accurately recognized, thereby improving the accuracy of recognizing control instructions in the voice data.

Description of drawings

Fig. 1 shows the flowchart of the speech recognition method according to some embodiments of the present application;

Fig. 2 shows a schematic diagram of a preset template provided by some embodiments of the present application;

Fig. 3 shows one of the schematic diagrams of the display interface provided by some embodiments of the present application;

Fig. 4 shows the second schematic diagram of the display interface provided by some embodiments of the present application;

Fig. 5 shows the third schematic diagram of the display interface provided by some embodiments of the present application;

FIG. 6 shows a structural block diagram of a speech recognition device according to an embodiment of the present application;

FIG. 7 shows a structural block diagram of an electronic device according to an embodiment of the present application;

FIG. 8 is a schematic diagram of a hardware structure of an electronic device implementing an embodiment of the present application.

Detailed ways

The following will clearly describe the technical solutions in the embodiments of the present application with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, but not all of them. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments in this application belong to the protection scope of this application.

The terms "first", "second" and the like in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific sequence or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the application are capable of operation in sequences other than those illustrated or described herein and that references to "first", "second", etc. to distinguish Objects are generally of one type, and the number of objects is not limited. For example, there may be one or more first objects. In addition, "and/or" in the specification and claims means at least one of the connected objects, and the character "/" generally means that the related objects are an "or" relationship.

The speech recognition method, device, electronic device and storage medium provided by the embodiments of the present application will be described in detail below through specific embodiments and application scenarios with reference to accompanying drawings 1 to 8 .

In some embodiments of the present application, a speech recognition method is provided. FIG. 1 shows a flow chart of a speech recognition method according to some embodiments of the present application. As shown in FIG. 1, the speech recognition method includes:

Step 102, receiving voice data, converting the voice data into first text data;

In the embodiment of the present application, the electronic device can receive voice data sent by the user, and the voice data is a voice command for the user to control the electronic device. After receiving the voice data, the electronic device converts the voice data into first text data through a voice recognition technology. The first text data corresponds to the voice data, and the first text data is a string of character fields.

Step 104, obtaining the pronunciation information of the first phrase data in the first text data;

In the embodiment of the present application, the first text data includes a plurality of phrase data, and each phrase data has a separate semantic meaning, and the electronic device can identify the first phrase data in the plurality of phrase data, and after recognizing the first phrase data In the case of , analyze the pronunciation information of the first phrase data. The pronunciation information is information such as pinyin corresponding to the first phrase data.

Specifically, the first phrase data can be a noun phrase, because the noun phrase in the voice command is usually a phrase that represents the control target in the voice command, for example: the voice command is "Please help me open the address book", where "address book" is for the first phrase.

Step 106, according to the pronunciation information and the first corresponding relationship, add an identification code to the first phrase data in the first text data to obtain the second text data;

Wherein, the first correspondence is the correspondence between pronunciation information and identification codes, and each identification code corresponds to at least two pronunciation information;

In the embodiment of the present application, the identification code is used to represent the pronunciation information of the first phrase data, so that the electronic device can accurately determine the corresponding intention of the voice data through the identification code. The electronic device is configured with a first correspondence, the first correspondence is the correspondence between the pronunciation information and the identification code, each identification code corresponds to multiple pronunciation information, and the multiple pronunciation information corresponding to the same identification code is similar pronunciation information , or confusing pronunciation information.

Exemplarily, the pinyin "nong", "long", and "rong" are set as the same identification code. "kang" and "kan" are set to the same identification code.

Step 108, based on the second text data, identify the control instruction corresponding to the voice data.

In the embodiment of the present application, the first text data generates the second text data by adding the identification code, that is, the second text data includes the pronunciation-related identification code, so as to improve the accuracy of the electronic device to determine the control instruction through the second text data.

In the embodiment of the present application, by adding an identification code to the first phrase data in the identified first text data, the electronic device can determine the true meaning of the first phrase data based on the identification code, so that even if the electronic device receives the first phrase data When the pronunciation of a phrase data is inaccurate, the true intention of the voice data including the first phrase data can still be accurately recognized, thereby improving the accuracy of recognizing control instructions in the voice data.

Exemplarily, when the voice data sent by the user is "Please help me open the communication channel", the electronic device generates the first text data "Please help me open the communication channel" based on the voice data, and recognizes that "Communication channel" is the first text data. As for the phrase data, an identification code is added to the first phrase data, because the identification code not only matches with "communication anger", but also matches with "address book". When identifying the semantics through the second text data, based on the semantics corresponding to the identification code, it can be accurately analyzed that the user's intention is to enable the address book function of the electronic device, so as to prevent the electronic device from being unable to recognize the user's inaccurate voice commands.

In the embodiment of the present application, when the electronic device receives the voice data, it converts the received voice data into the first text data, and adds an identification matching the pronunciation information to the first phrase data in the first text data encoding to generate the second text data, because the identification code in the second text data corresponds to a plurality of similar pronunciation information, so that the electronic device can determine the true meaning of the first phrase data based on the identification code, realizing that even if the electronic device receives Even when the pronunciation of the first phrase data is inaccurate, the true intention of the voice data including the first phrase data can still be accurately recognized, thereby improving the accuracy of recognizing control instructions in the voice data.

In some embodiments of the present application, obtaining the pronunciation information of the first phrase data in the first text data includes: performing word segmentation processing on the first text data to obtain at least two phrase data; The phrase data is determined as the first phrase data; and the pronunciation information of the first phrase data is acquired.

In the embodiment of the present application, after converting the voice data into the first text data, the electronic device performs word segmentation processing based on the semantics of the first text data to obtain multiple phrase data, that is, the multiple phrase data constitute the first text data. Identify the noun phrase data in the plurality of phrase data, determine the noun phrase data as the first phrase data, and acquire the pronunciation information of the first phrase data.

Specifically, the electronic device performs word segmentation processing on the first text data based on a template. For example, the template includes prefix word slots, verb word slots, and core word slots. The core word slots are usually noun phrases, and the The noun phrase refers to the control target in the corresponding control instruction. Wherein, the prefix word slot usually includes address words and auxiliary words, the verb word slot corresponds to the verb corresponding to the action performed in the first text data, and the core word word slot corresponds to the noun corresponding to the control target of the action performed in the first text data. For example: the first text data is "Please help me turn on the TV", and the word segmentation result is "Please help me/open/TV", wherein, "Please help me" is added to the prefix word slot, "Open" is added to the verb word slot, "Television" is added to the core word slot. The electronic device determines "television" added to the core word slot as the first phrase data, and adds a corresponding identification code to it, which can improve the accuracy of recognizing the corresponding control instruction of the voice data.

In some possible implementation manners, the verb phrase data and the noun phrase data in the at least two phrase data are determined as the first phrase data. By setting the verb phrase data as the first phrase data and adding corresponding identification codes, it is possible to improve the accuracy of the electronic device in recognizing the actions that the user needs to perform in the voice data, and further improve the accuracy of the electronic device in recognizing voice commands. sex.

In the embodiment of the present application, after the electronic device converts the voice data into the first text data, the first text data is word-segmented, and the noun phrase data in the first text data is used as the first phrase that affects the recognition accuracy of the voice data It is not necessary to add identification codes to the entire piece of first text data, which not only reduces the data processing capacity of the electronic equipment, but also improves the accuracy of the voice command recognition of the electronic equipment.

In some embodiments of the present application, the first correspondence includes a first sub-correspondence and a second sub-correspondence;

According to the pronunciation information and the first corresponding relationship, adding identification codes to the first phrase data in the first text data, including:

Obtain the pinyin information in the pronunciation information; according to the first sub-correspondence, find the first sub-code corresponding to the initial consonant information in the pinyin information, and each first sub-code corresponds to at least two initial consonant information; according to the second sub-correspondence, find The second sub-code corresponding to the final information in the pinyin information, each second sub-code corresponds to at least two finals information; according to the first sub-code and the second sub-code, generate an identification code, and add the identification code to the first text data.

In the embodiment of the present application, the pronunciation information corresponding to the first phrase data includes pinyin information, and the pinyin information includes initial consonant information and final consonant information.

In the embodiment of the present application, the identification code includes a first sub-code and a second sub-code, the first sub-code represents initial consonant information in the first phrase data, and the second sub-code represents final information in the first phrase data.

In the embodiment of the present application, the first correspondence is the correspondence between the first phrase data and the first coding identifier, the first correspondence includes the first sub-correspondence and the second sub-correspondence, and the first sub-correspondence is the first sub-correspondence The corresponding relationship between the code and the initial consonant information, and the second sub-correspondence is the corresponding relationship between the second sub-code and the final consonant information.

Specifically, the DimSim algorithm (a Chinese matching algorithm based on fuzzy sounds) is used to encode all initials and finals two-dimensionally, and the initials and finals are quantitatively identified by coordinate values (x, y). For example: the two-dimensional encoding of the initial "b" is (1.0,0.5), the encoding of the initial "P" is (1.0,1.5), the encoding of the final "a" is (1.0,0.0), and the encoding of the final "an" is (1.0,1.0). The two-dimensional codes of initials and finals are clustered by k-means clustering algorithm, and it can be determined that the initials "p" and "b" belong to the same class, and "h" and "f" belong to the same class.

The initials are grouped based on the above clustering algorithm, and the same first sub-code is set for the same group of initials, as shown in Table 1.

Table 1

group initials first subcode 1 p/b ~ 2 h/f ! 3 k/g @ 4 t/d # 5 r/l/n $ 6 z/zh/c/ch/j/q % 7 sh/s/x ^ 8 w w 9 the y the y 10 m m

The finals are grouped based on the above clustering algorithm, and the same second sub-coding is set for the same group of finals, as shown in Table 2.

group vowels representative character 1 ia/ian/iang/a/an/ang/ua/uan/uang & 2 ao/iao/ai/uai * 3 o/io ( 4 Iou/ou/uo ) 5 ong/iong - 6 ve/ie/ei/uei = 7 e/er/en/eng/uen/ueng _ 8 i/in/ing/v/vn +

In the embodiment of the present application, after the electronic device analyzes the first phrase data to obtain the pinyin information, it can find the first sub-code and the second sub-code respectively corresponding to the initial consonant information and the final consonant information in the pinyin information in the first phrase data , so as to generate the identification code corresponding to the first phrase data.

Exemplarily, the identification codes corresponding to the first phrase data "communication anger" and "address book" are both "#-^+$+".

Exemplarily, the first text data is "please open the communication rate", the first phrase data is "communication rate", and the second text data obtained by adding an identification code to the first phrase data in the first text data is " Please turn on the communication rate #-^+$+". The electronic device can match the "address book" through the identification code "#-^+$+" in the second text data, and execute the action of opening the address book of the electronic device.

In the embodiment of the present application, the electronic device finds the corresponding first sub-code and the second sub-code respectively through the initial consonant information and the final consonant information in the pinyin information of the first phrase data, and determines through the first sub-code and the second sub-code The identification code enables the electronic device to accurately identify the first phrase data with similar pronunciation.

In some embodiments of the present application, based on the recognition of the second text data, identifying the control instruction corresponding to the voice data includes: identifying the control intention information of the voice data according to the second phrase data in the second text data, the second phrase data Including the verb phrase data in the second text data; According to the first phrase data and the identification code in the second text data, determine the first application information of the voice data, the first application information is the information of the application program that executes the control instruction; according to Control the intent information and the first application information to generate a control command.

In the embodiment of the present application, the second phrase data is the verb phrase data in the second text data, and the electronic device can determine the control intention information required by the user to control the electronic device according to the second phrase data, and the control intention information is required by the electronic device Information about the action performed. The first application information is the control object in the control instruction. Since the first phrase data is noun phrase data, the electronic device can determine the control object controlled by the control instruction through the first phrase data.

Exemplarily, the second text data is "Please help me, please open the communication rate#-^+$+", wherein "open" is the second phrase data, and the electronic device can determine the control intention information through the second phrase data as " Open". "Communication rate #-^+$+" is the first phrase data, and the electronic device can determine that the first application information is "contact book" through the first phrase data.

Specifically, after the second text data is generated, the second text data is identified through a preset template, and when the second phrase data in the second text data is recognized, the control intention information is determined based on the semantics of the second phrase data , when the first phrase data in the second text data is recognized, the first application information is determined based on the semantics of the first phrase data. The electronic device can generate a corresponding control instruction based on the control intention information and the first application information, and execute the control instruction.

Figure 2 shows a schematic diagram of a preset template provided by some embodiments of the present application. As shown in Figure 2, the preset template includes a salutation word slot 202, an auxiliary word slot 204, a verb word slot 206, and a core word slot 208 And suffix word groove 210. Wherein, the salutation word slot 202 is used to identify the salutation words of the voice assistant in the electronic device, for example: "Little A", "Little V" and so on. The auxiliary word slot 204 is used to identify auxiliary words in the voice data, such as "please help me", "please help" and so on. The verb slot 206 is used to identify verb phrases in the speech data, such as "open", "open", "close" and so on. The core word slot 208 is used to identify noun phrases in the speech data, for example: "address book", "video software", "reading software" and so on. The suffix word slot 210 is used to identify modal particles such as "le" and "bar". When it is recognized that the second text data includes the first phrase data and the second phrase data, the corresponding control intention information and The first application information generates corresponding control instructions.

In the embodiment of the present application, after the electronic device converts the first text data into the second text data, the electronic device can determine the first application information based on the first phrase data in the second text data, and based on the second The phrase data determines the control intention information, and generates a control instruction according to the control intention information and the first application information.

In the embodiment of the present application, by extracting the key phrase data in the second text data and matching the corresponding control instructions of the second text data, the steps of obtaining the control instructions are simplified, and the second text data includes identification codes, which can improve Identify the accuracy and efficiency of control commands.

In some embodiments of the present application, the first application information of the voice data is determined according to the first phrase data and the identification code in the second text data, including:

In the case that at least two second application information corresponding to the identification code are found through the first phrase data, at least two application identifications are displayed, and at least two application identifications are in one-to-one correspondence with at least two second application information; in response to For the first input of the target application ID among the at least two application IDs, first application information corresponding to the target application ID is determined, and the first application information is one of the second application information.

In the embodiment of the present application, the second application information includes the name of the application program, and the electronic device searches for the corresponding second application information based on the identification code of the first phrase data. In the case that multiple different application programs have names with similar pronunciations, Then the identification codes corresponding to the multiple application programs are the same, so multiple second application information may be found. When the electronic device finds a plurality of second application information, it displays the corresponding plurality of application identifiers on the display screen, and the user selects the corresponding application identifier by performing a first input on the target application identifier, thereby determining the plurality of second application identifiers. The first application information in the information.

Fig. 3 shows one of the schematic diagrams of the display interface provided by some embodiments of the present application. As shown in Fig. 3, when the electronic device receives voice data as "open qi car application", the electronic device finds "car application" and Two application programs are "bicycle application", and a "car application" logo 302 and a "bicycle application" logo 304 are displayed on the display interface.

In this embodiment of the present application, the found second application information may be local second application information, or second application information in an application store. For example, when the electronic device receives the voice data as "open the qi car application", the electronic device detects that the relevant application program is not installed locally, then the electronic device finds a plurality of second application information in the application store, and then in The display interface displays the corresponding application identification and provides a download channel. The user can select the first application information among the plurality of second application information by performing the first input on the application identifiers of the plurality of second application information.

Fig. 4 shows the second schematic diagram of the display interface provided by some embodiments of the present application. As shown in Fig. 4, when the electronic device receives the voice data as "open the qi car application", the electronic device detects that the qi car application is not installed locally. The voice data matches the application program, and the electronic device finds two applications, "automobile application" and "bicycle application" in the application store, and the electronic device displays the "automobile application" logo 402 on the display interface, and the "bicycle application" ” logo 404, and a “download” control 406 and a “details” control 408 are displayed at corresponding positions. The user can download the corresponding application program by clicking the “Download” control 406 , and the user can view the detailed information of the corresponding application program by clicking the “Details” control 408 .

In the embodiment of the present application, when a plurality of application programs whose names have similar pronunciations are identified through the second text data, the application identifications of the corresponding application programs are displayed on the display screen for the user to select, so as to prevent the electronic device from accidentally triggering the corresponding application program. action command.

In some embodiments of the present application, according to the first phrase data and the identification code in the second text data, determining the first application information of the speech data includes: when at least two second application information are found through the first phrase data and the identification code In the case of two application information, determine the running priority of at least two second application information, and the running priority is associated with the historical running records of at least two second application information; based on the running priority, determine at least two second applications The first application information in the information.

In the embodiment of the present application, when the electronic device finds a plurality of second application information, the electronic device sorts the execution priorities of the plurality of second application information, and executes the application program corresponding to the second application information with the highest priority. corresponding control instructions.

In the embodiment of the present application, the running priority of the second application information is related to the historical running record of the application program, and the historical running record includes but not limited to the time, place, frequency, etc. of running the application program.

For example, when the electronic device finds two application programs through voice data, it displays the application identifications of the two application programs. After the user selects the target application identification from the two application identifications by clicking, the electronic device displays a prompt message "Voice Assistant "Always open the application selected this time" option, when the user selects this option, record this running record, and when the electronic device finds the above two applications again based on the voice data, it will always use the previously recorded application as the target application.

Fig. 5 shows the third schematic diagram of the display interface provided by some embodiments of the present application. As shown in Fig. 5, when the electronic device finds application A and application B according to the voice data, the display interface displays the corresponding The first identifier 502, and the second identifier 504 corresponding to the application B. After the user clicks on the first logo 502, the display interface displays the tab 506 of "the voice assistant always opens the application selected this time". After the user clicks on the tab 506, the electronic device records this selection, and finds it through the voice data next time. In the case of application A and application B, the electronic device always selects application A corresponding to the first identifier 502 to execute the corresponding control instruction.

In the embodiment of the present application, the electronic device can collect statistics on the historical records of different application programs used by the user.

Exemplarily, the number of times users use applications with similar pronunciations in each segment is counted, and the application with the highest frequency is selected to be associated with this time period. For example: in the usage records, from 8:00 to 8:10 in the morning, the number of openings of application A is 0, and the number of openings of application B is 2, then the application matching the time period of 8:00 to 8:10 in the morning is application B . When the electronic device finds the application A and the application B through the voice data during the period of 8:00-8:10 in the morning, the electronic device automatically starts the application B.

For example, the number of times users use applications with similar pronunciations in different locations is counted, and the application with the highest frequency is selected to be associated with the location. For example: Divide the places of use into shopping malls, communities, supermarkets, highways, subway stations, companies, scenic spots, etc., and count the number of uses of application A and application B in each location. In the usage records, the number of times application A is opened in the company is 10 times, the number of times that application B has been opened in the company is 0, and the company is associated with application A. The electronic device automatically starts the application A when the company finds the application A and the application B through the voice data.

In the embodiment of the present application, the electronic device can make statistics on the historical operation records of different application programs, and when multiple application programs with similar pronunciations are found through the voice data, the application program that executes the control command is automatically selected based on the historical operation records, Make electronic devices execute control instructions based on voice data to match the usage habits of users.

In some embodiments of the present application, a speech recognition device is provided. FIG. 6 shows a structural block diagram of a speech recognition device according to an embodiment of the present application. As shown in FIG. 6 , the speech recognition device 600 includes:

A processing module 602, configured to receive voice data, and convert the voice data into first text data;

An acquisition module 604, configured to acquire the pronunciation information of the first phrase data in the first text data;

The processing module 602 is configured to add an identification code to the first phrase data in the first text data according to the pronunciation information and the first correspondence to obtain second text data, wherein the first correspondence is the correspondence between the pronunciation information and the identification code Relationship, each identification code corresponds to at least two pronunciation information;

The processing module 602 is configured to recognize a control instruction corresponding to the voice data based on the second text data.

In the embodiment of the present application, by adding an identification code to the first phrase data in the identified first text data, the electronic device can determine the true meaning of the first phrase data based on the identification code, so that even if the electronic device receives the first phrase data When the pronunciation of a phrase data is inaccurate, the true intention of the voice data including the first phrase data can still be accurately recognized, thereby improving the accuracy of recognizing control instructions in the voice data.

In the embodiment of the present application, when the electronic device receives the voice data, it converts the received voice data into the first text data, and adds an identification matching the pronunciation information to the first phrase data in the first text data encoding to generate the second text data, because the identification code in the second text data corresponds to a plurality of similar pronunciation information, so that the electronic device can determine the true meaning of the first phrase data based on the identification code, realizing that even if the electronic device receives Even when the pronunciation of the first phrase data is inaccurate, the true intention of the voice data including the first phrase data can still be accurately recognized, thereby improving the accuracy of recognizing control instructions in the voice data.

In some embodiments of the present application, the processing module 602 is configured to perform word segmentation processing on the first text data to obtain at least two phrase data;

A processing module 602, configured to determine the noun phrase data in the at least two phrase data as the first phrase data;

An acquisition module 604, configured to acquire the pronunciation information of the first phrase data.

In the embodiment of the present application, after the electronic device converts the voice data into the first text data, the first text data is word-segmented, and the noun phrase data in the first text data is used as the first phrase that affects the recognition accuracy of the voice data It is not necessary to add identification codes to the entire piece of first text data, which not only reduces the data processing capacity of the electronic equipment, but also improves the accuracy of the voice command recognition of the electronic equipment.

In some embodiments of the present application, the first correspondence includes a first sub-correspondence and a second sub-correspondence;

An acquisition module 604, configured to acquire pinyin information in the pronunciation information;

The processing module 602 is used to search for the first sub-code corresponding to the initial consonant information in the pinyin information according to the first sub-correspondence, and each first sub-code corresponds to at least two initial consonant information;

The processing module 602 is used to search for the second sub-code corresponding to the final information in the pinyin information according to the second sub-correspondence, and each second sub-code corresponds to at least two finals information;

The processing module 602 is configured to generate an identification code according to the first sub-code and the second sub-code, and add the identification code to the first text data.

In the embodiment of the present application, the electronic device finds the corresponding first sub-code and the second sub-code respectively through the initial consonant information and the final consonant information in the pinyin information of the first phrase data, and determines through the first sub-code and the second sub-code The identification code enables the electronic device to accurately identify the first phrase data with similar pronunciation.

In some embodiments of the present application, the processing module 602 is configured to identify the control intention information of the speech data according to the second phrase data in the second text data, where the second phrase data includes verb phrase data in the second text data;

The processing module 602 is used to determine the first application information of the voice data according to the first phrase data and the identification code in the second text data, where the first application information is the information of the application program that executes the control instruction;

The processing module 602 is configured to generate a control instruction according to the control intention information and the first application information.

In the embodiment of the present application, after the electronic device converts the first text data into the second text data, the electronic device can determine the first application information based on the first phrase data in the second text data, and based on the second The phrase data determines the control intention information, and generates a control instruction according to the control intention information and the first application information.

In the embodiment of the present application, by extracting the key phrase data in the second text data and matching the corresponding control instructions of the second text data, the steps of obtaining the control instructions are simplified, and the second text data includes identification codes, which can improve Identify the accuracy and efficiency of control commands.

In some embodiments of the present application, the processing module 602 includes:

The display submodule is used to display at least two application identifications when at least two second application information corresponding to the identification code are found through the first phrase data, at least two application identifications and at least two second application information one-to-one correspondence;

The processing submodule is configured to determine first application information corresponding to the target application identifier in response to the first input of the target application identifier among the at least two application identifiers, where the first application information is one of the second application information.

In the embodiment of the present application, when a plurality of application programs whose names have similar pronunciations are identified through the second text data, the application identifications of the corresponding application programs are displayed on the display screen for the user to select, so as to prevent the electronic device from accidentally triggering the corresponding application program. action command.

In some embodiments of the present application, the processing module 602 is configured to determine the running priority of the at least two second application information when at least two second application information are found through the first phrase data and the identification code, The running priority is associated with at least two historical running records of the second application information;

The processing module 602 is configured to determine the first application information among the at least two second application information based on the running priority.

In the embodiment of the present application, the electronic device can make statistics on the historical operation records of different application programs, and when multiple application programs with similar pronunciations are found through the voice data, the application program that executes the control command is automatically selected based on the historical operation records, Make electronic devices execute control instructions based on voice data to match the usage habits of users.

The speech recognition apparatus in the embodiment of the present application may be an electronic device, or may be a component in the electronic device, such as an integrated circuit or a chip. The electronic device may be a terminal, or other devices other than the terminal. Exemplarily, the electronic device may be a mobile phone, a tablet computer, a notebook computer, a handheld computer, a vehicle electronic device, a mobile Internet device (Mobile Internet Device, MID), an augmented reality (augmented reality, AR)/virtual reality (virtual reality, VR) Devices, robots, wearable devices, ultra-mobile personal computers (ultra-mobile personalcomputer, UMPC), netbooks or personal digital assistants (personal digital assistant, PDA), etc., can also serve as servers, network attached storage (Network Attached Storage, NAS) , a personal computer (personal computer, PC), a television (television, TV), a teller machine or a self-service machine, etc., which are not specifically limited in this embodiment of the present application.

The voice recognition device in the embodiment of the present application may be a device with an operating system. The operating system may be an Android operating system, an iOS operating system, or other possible operating systems, which are not specifically limited in this embodiment of the present application.

The speech recognition device provided by the embodiment of the present application can realize each process realized by the above method embodiment, and to avoid repetition, details are not repeated here.

Optionally, the embodiment of the present application also provides an electronic device. FIG. 7 shows a structural block diagram of the electronic device according to the embodiment of the present application. As shown in FIG. 7 , the electronic device 700 includes a processor 702, a memory 704, a storage The programs or instructions that are on the memory 704 and can run on the processor 702, when the program or instructions are executed by the processor 702, implement the various processes of the above-mentioned method embodiments, and can achieve the same technical effect. To avoid repetition, they are not described here Let me repeat.

It should be noted that the electronic devices in the embodiments of the present application include the above-mentioned mobile electronic devices and non-mobile electronic devices.

FIG. 8 is a schematic diagram of a hardware structure of an electronic device implementing an embodiment of the present application.

The electronic device 800 includes, but is not limited to: a radio frequency unit 801, a network module 802, an audio output unit 803, an input unit 804, a sensor 805, a display unit 806, a user input unit 807, an interface unit 808, a memory 809, and a processor 810, etc. .

Those skilled in the art can understand that the electronic device 800 can also include a power supply (such as a battery) for supplying power to various components, and the power supply can be logically connected to the processor 810 through the power management system, so that the management of charging, discharging, and function can be realized through the power management system. Consumption management and other functions. The structure of the electronic device shown in FIG. 8 does not constitute a limitation to the electronic device. The electronic device may include more or fewer components than shown in the figure, or combine some components, or arrange different components, and details will not be repeated here. .

Wherein, the processor 810 is configured to receive voice data and convert the voice data into first text data;

Processor 810, configured to acquire the pronunciation information of the first phrase data in the first text data;

The processor 810 is configured to add an identification code to the first phrase data in the first text data according to the pronunciation information and the first correspondence to obtain second text data, wherein the first correspondence is the correspondence between the pronunciation information and the identification code Relationship, each identification code corresponds to at least two pronunciation information;

The processor 810 is configured to recognize a control instruction corresponding to the voice data based on the second text data.

In the embodiment of the present application, by adding an identification code to the first phrase data in the identified first text data, the electronic device can determine the true meaning of the first phrase data based on the identification code, so that even if the electronic device receives the first phrase data When the pronunciation of a phrase data is inaccurate, the true intention of the voice data including the first phrase data can still be accurately recognized, thereby improving the accuracy of recognizing control instructions in the voice data.

In the embodiment of the present application, when the electronic device receives the voice data, it converts the received voice data into the first text data, and adds an identification matching the pronunciation information to the first phrase data in the first text data encoding to generate the second text data, because the identification code in the second text data corresponds to a plurality of similar pronunciation information, so that the electronic device can determine the true meaning of the first phrase data based on the identification code, realizing that even if the electronic device receives Even when the pronunciation of the first phrase data is inaccurate, the true intention of the voice data including the first phrase data can still be accurately recognized, thereby improving the accuracy of recognizing control instructions in the voice data.

Further, the processor 810 is configured to perform word segmentation processing on the first text data to obtain at least two phrase data;

Processor 810, configured to determine the noun phrase data in the at least two phrase data as the first phrase data;

Processor 810, configured to acquire pronunciation information of the first phrase data.

In the embodiment of the present application, after the electronic device converts the voice data into the first text data, the first text data is word-segmented, and the noun phrase data in the first text data is used as the first phrase that affects the recognition accuracy of the voice data It is not necessary to add identification codes to the entire piece of first text data, which not only reduces the data processing capacity of the electronic equipment, but also improves the accuracy of the voice command recognition of the electronic equipment.

Further, the first correspondence includes a first sub-correspondence and a second sub-correspondence;

Processor 810, configured to acquire pinyin information in the pronunciation information;

The processor 810 is configured to search for the first sub-code corresponding to the initial consonant information in the pinyin information according to the first sub-correspondence, and each first sub-code corresponds to at least two initial consonant information;

The processor 810 is configured to search for a second sub-code corresponding to the final information in the pinyin information according to the second sub-correspondence, and each second sub-code corresponds to at least two finals information;

The processor 810 is configured to generate an identification code according to the first sub-code and the second sub-code, and add the identification code to the first text data.

In the embodiment of the present application, the electronic device finds the corresponding first sub-code and the second sub-code respectively through the initial consonant information and the final consonant information in the pinyin information of the first phrase data, and determines through the first sub-code and the second sub-code The identification code enables the electronic device to accurately identify the first phrase data with similar pronunciation.

Further, the processor 810 is configured to identify the control intention information of the voice data according to the second phrase data in the second text data, where the second phrase data includes verb phrase data in the second text data;

The processor 810 is configured to determine the first application information of the voice data according to the first phrase data and the identification code in the second text data, where the first application information is the information of the application program that executes the control instruction;

The processor 810 is configured to generate a control instruction according to the control intention information and the first application information.

In the embodiment of the present application, after the electronic device converts the first text data into the second text data, the electronic device can determine the first application information based on the first phrase data in the second text data, and based on the second The phrase data determines the control intention information, and generates a control instruction according to the control intention information and the first application information.

In the embodiment of the present application, by extracting the key phrase data in the second text data and matching the corresponding control instructions of the second text data, the steps of obtaining the control instructions are simplified, and the second text data includes identification codes, which can improve Identify the accuracy and efficiency of control commands.

Further, the display unit 806 is configured to display at least two application identifications when at least two second application information corresponding to the identification codes are found through the first phrase data, at least two application identifications and at least two second application information One-to-one correspondence between two application information;

The processor 810 is configured to determine first application information corresponding to the target application identifier in response to a first input of the target application identifier among the at least two application identifiers, where the first application information is one of the second application information.

In the embodiment of the present application, when a plurality of application programs whose names have similar pronunciations are identified through the second text data, the application identifications of the corresponding application programs are displayed on the display screen for the user to select, so as to prevent the electronic device from accidentally triggering the corresponding application program. action command.

Further, the processor 810 is configured to determine the running priority of the at least two second application information when at least two second application information are found through the first phrase data and the identification code, and the running priority is the same as the at least two second application information. Associated with the historical operation record of the second application information;

The processor 810 is configured to determine the first application information among the at least two second application information based on the running priority.

In the embodiment of the present application, the electronic device can make statistics on the historical operation records of different application programs, and when multiple application programs with similar pronunciations are found through the voice data, the application program that executes the control command is automatically selected based on the historical operation records, Make electronic devices execute control instructions based on voice data to match the usage habits of users.

It should be understood that, in the embodiment of the present application, the input unit 804 may include a graphics processing unit (Graphics Processing Unit, GPU) 8041 and a microphone 8042, and the graphics processing unit 8041 is compatible with an image capture device (such as Camera) to process the image data of still pictures or videos. The display unit 806 may include a display panel 8061, and the display panel 8061 may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like. The user input unit 807 includes at least one of a touch panel 8071 and other input devices 8072 . The touch panel 8071 is also called a touch screen. The touch panel 8071 may include two parts, a touch detection device and a touch controller. Other input devices 8072 may include, but are not limited to, physical keyboards, function keys (such as volume control buttons, switch buttons, etc.), trackballs, mice, and joysticks, which will not be repeated here.

The memory 809 can be used to store software programs as well as various data. The memory 809 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instructions required by at least one function (such as a sound playing function, image playback function, etc.), etc. Furthermore, memory 809 may include volatile memory or nonvolatile memory, or, memory 809 may include both volatile and nonvolatile memory. Among them, the non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electronically programmable Erase Programmable Read-Only Memory (Electrically EPROM, EEPROM) or Flash. Volatile memory can be random access memory (Random Access Memory, RAM), static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous connection dynamic random access memory (Synch link DRAM , SLDRAM) and Direct Memory Bus Random Access Memory (Direct Rambus RAM, DRRAM). The memory 809 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.

The processor 810 may include one or more processing units; optionally, the processor 810 integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to the operating system, user interface, and application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the foregoing modem processor may not be integrated into the processor 810 .

The embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored, and when the program or instruction is executed by the processor, each process of the above-mentioned method embodiment can be realized, and the same technical effect can be achieved. To avoid repetition, details are not repeated here.

Wherein, the processor is the processor in the electronic device in the foregoing embodiments. A readable storage medium includes a computer-readable storage medium, such as a computer read-only memory (Read-Only Memory, ROM), a random access memory (Random Access Memory, RAM), a magnetic disk or an optical disk, and the like.

The embodiment of the present application further provides a chip, the chip includes a processor and a communication interface, the communication interface is coupled to the processor, the processor is used to run programs or instructions, and realize the various processes of the above method embodiments, and can achieve the same technical effect , to avoid repetition, it will not be repeated here.

It should be understood that the chips mentioned in the embodiments of the present application may also be called system-on-chip, system-on-chip, system-on-a-chip, or system-on-a-chip.

The embodiment of the present application provides a computer program product, the program product is stored in a storage medium, and the program product is executed by at least one processor to implement the various processes in the above method embodiments, and can achieve the same technical effect, for To avoid repetition, I won't go into details here.

It should be noted that, in this document, the term "comprising", "comprising" or any other variation thereof is intended to cover a non-exclusive inclusion such that a process, method, article or apparatus comprising a set of elements includes not only those elements, It also includes other elements not expressly listed, or elements inherent in the process, method, article, or device. Without further limitations, an element defined by the phrase "comprising a ..." does not preclude the presence of additional identical elements in the process, method, article, or apparatus comprising that element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved. Functions are performed, for example, the described methods may be performed in an order different from that described, and various steps may also be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

Through the description of the above embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus a necessary general-purpose hardware platform, and of course also by hardware, but in many cases the former is better implementation. Based on such an understanding, the technical solution of the present application can be embodied in the form of computer software products, which are stored in a storage medium (such as ROM/RAM, magnetic disk, etc.) , optical disc), including several instructions to enable a terminal (which may be a mobile phone, computer, server, or network device, etc.) to execute the methods of various embodiments of the present application.

The embodiments of the present application have been described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementations. The above-mentioned specific implementations are only illustrative and not restrictive. Those of ordinary skill in the art will Under the inspiration of this application, without departing from the purpose of this application and the scope of protection of the claims, many forms can also be made, all of which belong to the protection of this application.

Claims (14)

1. A method of speech recognition, comprising:

receiving voice data, and converting the voice data into first text data;

acquiring pronunciation information of first phrase data in the first text data;

adding identification codes to the first phrase data in the first text data according to the pronunciation information and a first corresponding relation to obtain second text data, wherein the first corresponding relation is the corresponding relation between the pronunciation information and the identification codes, and each identification code corresponds to at least two pronunciation information;

and identifying a control instruction corresponding to the voice data based on the second text data.

2. The method of claim 1, wherein the obtaining the pronunciation information of the first phrase data in the first text data includes:

word segmentation processing is carried out on the first text data to obtain at least two phrase data;

determining noun phrase data in the at least two phrase data as the first phrase data;

and acquiring pronunciation information of the first phrase data.

3. The method of claim 1, wherein the first correspondence includes a first sub-correspondence and a second sub-correspondence;

The adding identification codes to the first phrase data in the first text data according to the pronunciation information and the first corresponding relation comprises the following steps:

acquiring pinyin information in the pronunciation information;

searching first subcodes corresponding to the initial consonant information in the pinyin information according to the first subcodes corresponding relation, wherein each first subcode corresponds to at least two initial consonant information;

searching second subcodes corresponding to the vowel information in the pinyin information according to the second subcode corresponding relation, wherein each second subcode corresponds to at least two vowel information;

and generating the identification code according to the first sub-code and the second sub-code, and adding the identification code into the first text data.

4. A speech recognition method according to any one of claims 1 to 3, wherein said recognizing the control instruction corresponding to the speech data based on the second text data recognition comprises:

identifying control intention information of the voice data according to second phrase data in the second text data, wherein the second phrase data comprises verb phrase data in the second text data;

Determining first application information of the voice data according to the first phrase data and the identification code in the second text data, wherein the first application information is information of an application program executing the control instruction;

and generating the control instruction according to the control intention information and the first application information.

5. The method according to claim 4, wherein said determining the first application information of the speech data based on the first phrase data and the identification code in the second text data includes:

displaying at least two application identifiers under the condition that at least two pieces of second application information corresponding to the identifier codes are searched through the first phrase data, wherein the at least two application identifiers are in one-to-one correspondence with the at least two pieces of second application information;

and responding to the first input of a target application identifier in the at least two application identifiers, and determining first application information corresponding to the target application identifier, wherein the first application information is one of the second application information.

6. The method according to claim 4, wherein said determining the first application information of the speech data based on the first phrase data and the identification code in the second text data includes:

Determining the operation priority of at least two second application information under the condition that the at least two second application information is found through the first phrase data and the identification code, wherein the operation priority is associated with historical operation records in the at least two second application information;

and determining the first application information in the at least two second application information based on the operation priority.

7. A speech recognition apparatus, comprising:

the processing module is used for receiving voice data and converting the voice data into first text data;

the acquisition module is used for acquiring pronunciation information of the first phrase data in the first text data;

the processing module is used for adding identification codes to the first phrase data in the first text data according to the pronunciation information and a first corresponding relation to obtain second text data, wherein the first corresponding relation is a corresponding relation between the pronunciation information and the identification codes, and each identification code corresponds to at least two pronunciation information;

the processing module is used for identifying the control instruction corresponding to the voice data based on the second text data.

8. The speech recognition apparatus of claim 7, wherein,

the processing module is used for carrying out word segmentation processing on the first text data to obtain at least two phrase data;

the processing module is configured to determine noun phrase data in the at least two phrase data as the first phrase data;

the acquisition module is used for acquiring the pronunciation information of the first phrase data.

9. The speech recognition device of claim 7, wherein the first correspondence comprises a first sub-correspondence and a second sub-correspondence;

the acquisition module is used for acquiring pinyin information in the pronunciation information;

the processing module is used for searching first subcodes corresponding to the initial consonant information in the pinyin information according to the first subcodes corresponding relation, and each first subcode corresponds to at least two initial consonant information;

the processing module is used for searching second subcodes corresponding to the vowel information in the pinyin information according to the second subcode corresponding relation, and each second subcode corresponds to at least two vowel information;

the processing module is used for generating the identification code according to the first sub-code and the second sub-code and adding the identification code into the first text data.

10. The voice recognition device according to any one of claims 7 to 9,

the processing module is used for identifying control intention information of the voice data according to second phrase data in the second text data, wherein the second phrase data comprises verb phrase data in the second text data;

the processing module is used for determining first application information of the voice data according to the first phrase data and the identification code in the second text data, wherein the first application information is information of an application program executing the control instruction;

the processing module is used for generating the control instruction according to the control intention information and the first application information.

11. The speech recognition device of claim 10, wherein the processing module comprises:

the display sub-module is used for displaying at least two application identifiers under the condition that at least two second application information are found through the first phrase data, and the at least two application identifiers are in one-to-one correspondence with the at least two second application information;

and the processing sub-module is used for responding to the first input of the target application identifier in the at least two application identifiers and determining the second application information corresponding to the target application identifier as the first application information.

12. The speech recognition apparatus of claim 10, wherein,

the processing module is used for determining the operation priority of the at least two second application information under the condition that the at least two second application information is found out through the first phrase data, and the operation priority is associated with the historical operation records of the at least two second application information;

the processing module is configured to determine the first application information in the at least two second application information based on the operation priority.

13. An electronic device, comprising:

a processor and a memory storing a program or instructions executable on the processor, which when executed by the processor, implement the steps of the speech recognition method according to any one of claims 1 to 6.

14. A readable storage medium having stored thereon a program or instructions which when executed by a processor realizes the speech recognition method steps of any of claims 1 to 6.

Images