Voice processing method, device, electronic device and storage medium

Abstract

The embodiment of the present invention discloses a voice processing method, device, electronic device and storage medium. The method comprises: after playing the guidance information, obtaining the first voice information input by the user, wherein the guidance information is generated based on the second voice information input by the user; if the first voice information matches the guidance information, obtaining the first voice analysis result according to the first voice information, wherein the second voice information is associated with the first voice analysis result; and playing the first feedback information according to the first voice analysis result. Through the scheme in the embodiment of the present invention, it is possible to dynamically guide the user to input voice information, thereby improving the working efficiency of voice processing; associating the incomprehensible voice information with the understandable voice information result, thereby realizing online learning, which can improve the accuracy of the voice processing result and further improve the user experience.

Description

Technical Field

[0001] The embodiments of the present invention relate to natural language processing technology, and in particular to a speech processing method, device, electronic device and storage medium. Background Art

[0002] With the continuous development of the field of intelligent question and answer, voice assistants are increasingly used in smart homes, smart cars, smart customer service and other fields.

[0003] Existing voice assistants often fail to understand user commands. When faced with such situations, voice assistants generally use small talk or blunt replies such as "I'm sorry, I didn't understand" to reject user commands. This results in voice assistants being less intelligent, greatly affecting user experience and the completion rate of user tasks. Summary of the invention

[0004] The embodiments of the present invention provide a speech processing method, device, electronic device and storage medium, which can improve the intelligence level of speech processing and enhance user experience.

[0005] In a first aspect, an embodiment of the present invention provides a speech processing method, comprising:

[0006] After playing the guidance information, obtaining first voice information input by the user, wherein the guidance information is generated based on second voice information input by the user;

[0007] If the first voice information matches the guide information, obtaining a first voice analysis result according to the first voice information, wherein the second voice information is associated with the first voice analysis result;

[0008] Play first feedback information according to the first speech analysis result.

[0009] In a second aspect, an embodiment of the present invention provides a speech processing device, the device comprising:

[0010] An acquisition module, used for acquiring first voice information input by a user after playing the guidance information, wherein the guidance information is generated based on second voice information input by the user;

[0011] a parsing module, configured to obtain a first voice parsing result according to the first voice information if the first voice information matches the guide information, wherein the second voice information is associated with the first voice parsing result;

[0012] A feedback module is used to play first feedback information according to the first speech analysis result.

[0013] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a speech processing method as described in any one of the embodiments of the present invention is implemented.

[0014] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a speech processing method as described in any one of the embodiments of the present invention.

[0015] In the embodiment of the present invention, based on the second voice information input by the user, the guidance information is played for the user, the first voice information input by the user is obtained, and if the first voice information matches the guidance information, the first voice analysis result is obtained according to the first voice information, wherein the second voice information is associated with the first voice analysis result; and according to the first voice analysis result, the first feedback information is played. Through this solution, it is possible to dynamically guide the user to input voice information, thereby improving the working efficiency of voice processing; incomprehensible voice information is associated with understandable voice information results, thereby realizing online learning, which can improve the accuracy of voice processing results and further improve user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0017] Figure 1 A flow chart of a speech processing method provided by an embodiment of the present invention;

[0018] Figure 2 Another flowchart of a speech processing method provided by an embodiment of the present invention;

[0019] Figure 3 is a structural diagram of a speech processing device provided by an embodiment of the present invention;

[0020] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0021] The present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It is to be understood that the specific embodiments described herein are only used to explain the present invention, rather than to limit the present invention. It should also be noted that, for ease of description, only parts related to the present invention, rather than all structures, are shown in the accompanying drawings.

[0022] Figure 1 The present invention provides a flow chart of a speech processing method. The method of the present invention can be applied to intelligent question-answering scenarios. The method can be executed by a speech processing device provided by the present invention. The device can be implemented in software and / or hardware. In a specific embodiment, the device can be integrated into an electronic device, which can be a computer, a server, etc. The following embodiments will be described by taking the device integrated into an electronic device as an example. Figure 1 , the method may specifically include the following steps:

[0023] Step 101: After playing the guidance information, obtaining first voice information input by the user, wherein the guidance information is generated based on second voice information input by the user.

[0024] Among them, the second voice information input by the user can be voice instructions, questions or chats sent by the user based on his own needs. After receiving the second voice information sent by the user, the second voice information can be converted into text information through the voice recognition system for subsequent processing. Of course, the user can also directly input the second voice information in text form through the device. After receiving the second voice information, the second voice information can be semantically parsed through the semantic understanding model, and further, guidance information can be generated based on the second voice information input by the user. Furthermore, the guidance information is played for the user, and the text information corresponding to the guidance information can also be displayed to the user. After receiving the guidance information, the user can input the first voice information according to the guidance information, and further, the first voice information input by the user is obtained.

[0025] For example, the user wants to close the windows through the in-car voice assistant, and the second voice message input by the user is "Close all the windows". After the voice assistant receives the second voice message input by the user, it performs semantic analysis on the second voice message and generates guidance information for the second voice message based on the analysis result. Assuming that the generated guidance information is "Close all windows", the guidance information is played for the customer: "You can say: Close all windows" or "Do you want to close all windows". The user can input the first user instruction according to the guidance information, such as "Close all windows" or "Yes". Further, the first voice message input by the user is obtained.

[0026] Step 102: If the first voice information matches the guidance information, a first voice analysis result is obtained according to the first voice information, wherein the second voice information is associated with the first voice analysis result.

[0027] Specifically, after receiving the first voice information input by the user, the first voice information is matched with the guidance information. If the first voice information matches the guidance information, the first voice analysis result is obtained according to the first voice information. Among them, the analysis result of the first voice information is the analysis result of the guidance information, and the second voice information is associated with the first voice analysis result. For example, the second voice information input by the user is "Close all the windows", assuming that the generated guidance information is "Close all the windows", and the first voice information input by the user is "Close all the windows". Through the semantic matching model, the first voice information input by the user and the guidance information are matched. According to the guidance information and the first voice information, it can be known that the first voice analysis result is that the user needs to close all the windows. Furthermore, by associating the second voice information with the first voice analysis result, it can be known that if the user inputs "Close all the windows", the analysis result is the analysis result of "Close all the windows".

[0028] Step 103: Play first feedback information according to the first voice analysis result.

[0029] Specifically, after the user inputs the second voice information, a guide message is generated for the user according to the second voice information, and the user is waiting to input the first voice information. After the user inputs the first voice information, if the first voice information matches the guide message, a first voice analysis result is generated according to the analysis result of the guide message and the first voice information, and the first voice analysis result is played or displayed in text for the user. The second voice information is associated with the first voice analysis result.

[0030] The technical solution of this embodiment is to obtain the first voice information input by the user after playing the guidance information. If the first voice information matches the guidance information, the first voice analysis result is obtained according to the first voice information, wherein the second voice information is associated with the first voice analysis result; and the first feedback information is played according to the first voice analysis result. This solution can dynamically guide the user to input voice information, thereby improving the working efficiency of voice processing; and associate the incomprehensible voice information with the understandable voice information results, thereby realizing online learning, which can improve the accuracy of the voice processing results and further improve the user experience.

[0031] Figure 2 This is another flow chart of a speech processing method provided by an embodiment of the present invention. This embodiment refines the steps of the speech processing method based on the above embodiment. Figure 2 As shown, the method of this embodiment specifically includes the following steps:

[0032] Step 201: Acquire second voice information input by a user, and parse the second voice information.

[0033] Specifically, the second voice information input by the user is received. Wherein, when the second voice information is voice, the voice information is converted into text information. Of course, the user can also directly input text as the second voice information. After receiving the second voice information, the second voice information is semantically parsed through the semantic understanding model.

[0034] Step 202: If the second voice information fails to be parsed, determine whether there is a third voice information in the first database, wherein the similarity between the third voice information and the second voice information is greater than or equal to a preset threshold. Further, if the second voice information fails to be parsed, determine whether there is a third voice information in the first database. Specifically, the first database can be constructed in real time based on the user interaction log, and the first database stores guidance information that can be correctly parsed by the semantic understanding system. As shown in Table 1 below:

[0035] Table 1

[0036] Boot information Parsing results Close the sunshade Intent: Close the sunshade Sunroof half closed Intent: Sunroof Control, Action: Off, Value: 50% Beijing Tomorrow Weather Intent: Weather, Location: Beijing, Time: Tomorrow

[0037] It should be noted that the guidance information and analysis results shown in Table 1 are only examples and do not constitute a final limitation on the actual guidance information and analysis results. In actual applications, the relevant data can be adjusted according to actual needs, and no specific limitation is made here.

[0038] Further, it is determined whether there is third voice information in the first database, wherein the similarity between the third voice information and the second voice information is greater than or equal to a preset threshold, as shown in the following Table 2. It should be noted that the data shown in Table 2 are only examples and do not constitute a final limitation on the actual data setting. In actual applications, the relevant data can also be adjusted according to actual needs, and no specific limitation is made here.

[0039] Table 2

[0040] Second voice message Third voice message Similarity scoring Close the screen Close the sunshade 0.9800629019737244 Close the screen Close the window 0.5422488117218018

[0041] For example, when the second voice information is "Close the screen window", the similarity threshold is set to 0.8, and the third voice information is the voice information whose similarity with the second voice information is greater than or equal to 0.8. Determine whether the third voice information exists in the first database. Assuming that "Close the sunshade" exists in the first database, and the similarity between "Close the sunshade" and "Close the screen window" is greater than 0.8, it can be determined that the third voice information exists in the first database. If there are multiple third voice information in database 1, the one with the highest similarity threshold is determined as the third voice information.

[0042] Among them, different thresholds can be set according to different types of second voice information, and information with a similarity greater than or equal to the preset threshold with the second voice information is used as the third voice information. When the second voice information input by the user is voice information of the type of query location, the similarity threshold can be adaptively increased. For example, if the second voice information is "Navigate to Keji Road", and "Navigate to Keji Road" is stored in the first database, if the similarity threshold is set low and "Navigate to Keji Road" is used as the third voice information, it will affect the customer experience.

[0043] Step 203: If the third voice information exists in the first database, generate guidance information according to the third voice information.

[0044] After determining that the third voice information exists in the first database, the third voice information with the highest similarity value in the first database is read, the third voice information is parsed through the semantic understanding model, and guidance information is generated according to the third voice information. Further, the guidance information is played for the user to guide the user.

[0045] Step 204: After playing the guidance information, obtain the first voice information input by the user.

[0046] After receiving the guidance information, the user can input the first voice information according to the guidance information, and further, the first voice information input by the user is obtained.

[0047] Step 205: If the first voice information matches the guidance information, a first voice analysis result is obtained according to the first voice information, wherein the second voice information is associated with the first voice analysis result.

[0048] The analysis result of the first voice information is the analysis result of the guidance information, and the second voice information is associated with the first voice analysis result.

[0049] Specifically, after the user inputs the second voice information, a guide message is generated for the user according to the second voice information, and the user is waiting to input the first voice information. After the user inputs the first voice information, if the first voice information matches the guide message, a first voice analysis result is generated according to the analysis result of the guide message and the first voice information, and the first voice analysis result is played or displayed in text for the user.

[0050] In the embodiment of this solution, optionally, the second voice information is associated with the first voice analysis result and stored in a second database.

[0051] Specifically, when the first voice information matches the guidance information, after obtaining the first voice analysis result according to the first voice information, the analysis result of the first voice information is associated with the voice analysis result of the second information. And the second voice information and the first voice analysis result can be stored in the second database. Among them, the second database can be used as training corpus for subsequent semantic understanding model upgrade iteration.

[0052] The second voice information is associated with the first voice analysis result and stored in the second database, which facilitates the subsequent upgrade and iteration of the semantic understanding model, can improve the accuracy of the semantic analysis result, and thus improve the user experience

[0053] If the first voice information does not match the guidance information, the second voice information input by the user is obtained, and the second voice information is semantically parsed through the semantic understanding model (executing step 201).

[0054] Step 206: Play the first feedback information according to the first voice analysis result.

[0055] By setting a similarity threshold to determine the third voice information and generating guidance information according to the third voice information, the accuracy of the result fed back to the user can be improved, further improving the user experience.

[0056] Step 207: If the third voice information does not exist in the first database, determine whether the second voice information is chat information.

[0057] After determining that the third voice information does not exist in the first database, it is further determined whether the second voice information is idle chat information. For example, the second voice information input by the user is "I am so bored", and there is no third voice information with a similarity greater than a preset threshold in the first database. Then the voice information "Do you want to chat" can be played to the user.

[0058] Step 208: If the second voice information is small talk information, the third feedback information is played, wherein the third feedback information is small talk words.

[0059] If it is determined that the second voice message input by the user is a chat message, the chat reply words are played for the user, such as "Let's chat then".

[0060] Step 209: If the second voice information is not small talk information, play the fourth feedback information, wherein the fourth feedback information is used to indicate that the voice understanding fails.

[0061] If it is determined that the second voice information input by the user is not a small talk message, and the third voice information does not exist in the first database, the user is fed back information indicating that the voice comprehension failed, such as "I didn't understand".

[0062] Through the above steps, when the user does not accept the guidance information, an exit strategy is adopted, the user's needs are fully understood, and chat services are provided to the user to improve the user experience.

[0063] Step 210: If the second voice information is analyzed successfully, the second voice analysis result is obtained, and the second feedback information is played according to the second voice analysis result.

[0064] For example, the second voice message input by the user is: "Close all windows", and the semantic understanding model parses the intention of the second voice message as "Close all windows" ("Close all windows" is stored in the first database, and the semantic understanding model can obtain the parsing result through voice analysis). Then the feedback message "OK, close all windows for you" is played to the user.

[0065] Step 211: associate the second voice information with the second voice analysis result and store them in the first database.

[0066] For example, the first database stores the analysis result of "close all windows". After the second voice message ("close all windows") is successfully analyzed, the second voice message is associated with the second voice analysis result and stored in the first database.

[0067] In the embodiment of this solution, optionally, the first database and the second database are constructed in real time based on the user interaction log.

[0068] By constructing the first database and the second database in real time based on the user interaction log, manual labeling is not required, which saves human resources and further improves the working efficiency of the speech processing method.

[0069] In this embodiment, the second voice information input by the user is obtained and parsed; if the second voice information fails to be parsed, the guidance information is generated and played according to the second voice information; if the second voice information is parsed successfully, the second voice parsing result is obtained, and the second feedback information is played according to the second voice parsing result; the second voice information is associated with the second voice parsing result and stored in the first database. For the incomprehensible second voice information, the third voice information that can be understood is matched, so as to realize dynamic guidance of the second voice information input by the user and improve the learning efficiency of the voice processing method; at the same time, the iterative learning of the voice processing method is realized to improve the accuracy of the voice processing result.

[0070] Figure 3 is a structural diagram of a speech processing device provided by an embodiment of the present invention, and the device is suitable for executing the speech processing method provided by an embodiment of the present invention. Figure 3 As shown, the device may specifically include:

[0071] The acquisition module 301 is used to acquire the first voice information input by the user after playing the guidance information, wherein the guidance information is generated based on the second voice information input by the user;

[0072] The analysis module 302 is used for obtaining a first voice analysis result according to the first voice information if the first voice information matches the guide information, wherein the second voice information is associated with the first voice analysis result;

[0073] The feedback module 303 is used to play the first feedback information according to the first speech analysis result.

[0074] Optionally, the acquisition module 301 is further configured to acquire the second voice information input by the user and parse the second voice information;

[0075] The parsing module 302 is further configured to generate and play the guiding information according to the second voice information if the parsing of the second voice information fails.

[0076] Optionally, the analysis module 302 is specifically used to determine whether there is third voice information in the first database, wherein the similarity between the third voice information and the second voice information is greater than or equal to a preset threshold; if there is third voice information in the first database, the guide information is generated according to the third voice information.

[0077] Optionally, the analysis module 302 is further used to obtain a second voice analysis result if the second voice information is successfully analyzed, and play second feedback information according to the second voice analysis result; and associate the second voice information with the second voice analysis result and store them in the first database.

[0078] Optionally, the analysis module 302 is also used to determine whether the second voice information is chat information if the third voice information does not exist in the first database; if the second voice information is chat information, play third feedback information, wherein the third feedback information is chat words; if the second voice information is not chat information, play fourth feedback information, wherein the fourth feedback information is used to indicate that voice understanding failed.

[0079] Optionally, the analysis module 302 is further configured to associate the second voice information with the first voice analysis result and store them in a second database.

[0080] Optionally, the first database and the second database are constructed in real time based on user interaction logs.

[0081] The speech processing device provided in the embodiment of the present invention can execute the speech processing method provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method. For the contents not described in detail in this embodiment, reference can be made to the description in any method embodiment of the present invention.

[0082] Figure 4 A schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. Figure 4 A block diagram of an exemplary electronic device 12 suitable for use in implementing embodiments of the present invention is shown. Figure 4 The electronic device 12 shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0083] like Figure 4 As shown, the electronic device 12 is in the form of a general purpose computing device. The components of the electronic device 12 may include, but are not limited to: one or more processors or processing units 16, a system memory 28, and a bus 18 that connects various system components (including the system memory 28 and the processing unit 16).

[0084] Bus 18 represents one or more of several types of bus structures, including a memory bus or memory controller, a peripheral bus, an accelerated graphics port, a processor or a local bus using any of a variety of bus architectures. By way of example, these architectures include, but are not limited to, an Industry Standard Architecture (ISA) bus, a Micro Channel Architecture (MAC) bus, an Enhanced ISA bus, a Video Electronics Standards Association (VESA) local bus, and a Peripheral Component Interconnect (PCI) bus.

[0085] The electronic device 12 typically includes a variety of computer system readable media. These media can be any available media that can be accessed by the electronic device 12, including volatile and non-volatile media, removable and non-removable media.

[0086] The system memory 28 may include computer system readable media in the form of volatile memory, such as random access memory (RAM) 30 and / or cache memory 32. The electronic device 12 may further include other removable / non-removable, volatile / non-volatile computer system storage media. By way of example only, the storage system 34 may be used to read and write non-removable, non-volatile magnetic media ( Figure 4 not shown, usually called a "hard drive"). Although Figure 4Not shown in the figure, a disk drive for reading and writing to a removable non-volatile disk (e.g., a "floppy disk"), and an optical disk drive for reading and writing to a removable non-volatile optical disk (e.g., a CD-ROM, a DVD-ROM, or other optical media) may be provided. In these cases, each drive may be connected to the bus 18 via one or more data medium interfaces. The memory 28 may include at least one program product having a set (e.g., at least one) of program modules that are configured to perform the functions of the various embodiments of the present invention.

[0087] A program / utility 40 having a set (at least one) of program modules 42 may be stored, for example, in the memory 28. Such program modules 42 include, but are not limited to, an operating system, one or more application programs, other program modules, and program data, each of which or some combination may include an implementation of a network environment. The program modules 42 generally perform the functions and / or methods of the embodiments described herein.

[0088] The electronic device 12 may also communicate with one or more external devices 14 (e.g., keyboard, pointing device, display 24, etc.), may also communicate with one or more devices that enable a user to interact with the electronic device 12, and / or communicate with any device that enables the electronic device 12 to communicate with one or more other computing devices (e.g., network card, modem, etc.). Such communication may be performed through an input / output (I / O) interface 22. In addition, the electronic device 12 in this embodiment, the display 24 does not exist as an independent individual, but is embedded in the mirror surface, and when the display surface of the display 24 is not displayed, the display surface of the display 24 and the mirror surface are visually integrated. In addition, the electronic device 12 may also communicate with one or more networks (e.g., local area network (LAN), wide area network (WAN) and / or public network, such as the Internet) through a network adapter 20. As shown in the figure, the network adapter 20 communicates with other modules of the electronic device 12 through a bus 18. It should be understood that although not shown in the figures, other hardware and / or software modules may be used in conjunction with the electronic device 12, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives, and data backup storage systems.

[0089] The processing unit 16 executes various functional applications and data processing by running the program stored in the system memory 28, for example, implementing a voice processing method provided in an embodiment of the present invention: after playing the guidance information, obtaining the first voice information input by the user, wherein the guidance information is generated based on the second voice information input by the user; if the first voice information matches the guidance information, obtaining the first voice analysis result according to the first voice information, wherein the second voice information is associated with the first voice analysis result; and playing the first feedback information according to the first voice analysis result.

[0090] The embodiment of the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, a speech processing method as provided in all the embodiments of the present invention is implemented: after playing the guidance information, obtaining the first voice information input by the user, wherein the guidance information is generated based on the second voice information input by the user; if the first voice information matches the guidance information, obtaining the first voice analysis result according to the first voice information, wherein the second voice information is associated with the first voice analysis result; and playing the first feedback information according to the first voice analysis result. The computer-readable medium may be a computer-readable signal medium or a computer-readable storage medium. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination thereof. More specific examples (non-exhaustive list) of computer-readable storage media include: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In this document, computer readable storage media may be any tangible media that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device.

[0091] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, which carry computer-readable program code. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than a computer-readable storage medium, which may send, propagate, or transmit a program for use by or in conjunction with an instruction execution system, apparatus, or device.

[0092] The program code embodied on the computer readable medium may be transmitted using any appropriate medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

[0093] Computer program code for performing the operations of the present invention may be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, C++, and conventional procedural programming languages ​​such as "C" or similar programming languages. The program code may be executed entirely on the user's computer, partially on the user's computer, as a separate software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or may be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0094] Note that the above are only preferred embodiments of the present invention and the technical principles used. Those skilled in the art will understand that the present invention is not limited to the specific embodiments herein, and that various obvious changes, readjustments and substitutions can be made by those skilled in the art without departing from the scope of protection of the present invention. Therefore, although the present invention has been described in more detail through the above embodiments, the present invention is not limited to the above embodiments, and may include more other equivalent embodiments without departing from the concept of the present invention, and the scope of the present invention is determined by the scope of the appended claims.

Claims

1. A speech processing method, characterized in that: include: After playing the guidance information, obtaining first voice information input by the user, wherein the guidance information is generated based on second voice information input by the user; If the first voice information matches the guide information, obtaining a first voice analysis result according to the first voice information, wherein the second voice information is associated with the first voice analysis result; Playing first feedback information according to the first voice analysis result; The method further comprises: Before playing the guidance information, obtaining the second voice information input by the user and parsing the second voice information; if the parsing of the second voice information fails, generating and playing the guidance information according to the second voice information; Generating the guide information includes: Determine whether there is third voice information in the first database, wherein the similarity between the third voice information and the second voice information is greater than or equal to a preset threshold; If the third voice information exists in the first database, the guide information is generated according to the third voice information; if the third voice information does not exist in the first database, whether the second voice information is chat information is determined; If the second voice information is small talk information, the third feedback information is played, wherein the third feedback information is small talk words; if the second voice information is not small talk information, the fourth feedback information is played, wherein the fourth feedback information is used to indicate that the voice understanding fails.

2. The speech processing method according to claim 1, characterized in that: Also includes: If the second voice information is parsed successfully, obtaining the second voice parsing result, and playing the second feedback information according to the second voice parsing result; The second voice information is associated with the second voice analysis result and stored in the first database.

3. The speech processing method according to any one of claims 1 to 2, characterized in that: Also includes: The second voice information is associated with the first voice analysis result and stored in a second database.

4. The speech processing method according to claim 3, characterized in that: The first database and the second database are constructed in real time based on user interaction logs.

5. A speech processing device, characterized in that: include: An acquisition module, used for acquiring first voice information input by a user after playing the guidance information, wherein the guidance information is generated based on second voice information input by the user; a parsing module, configured to obtain a first voice parsing result according to the first voice information if the first voice information matches the guide information, wherein the second voice information is associated with the first voice parsing result; A feedback module, configured to play first feedback information according to the first speech analysis result; The acquisition module is further used to acquire the second voice information input by the user before playing the guidance information, and parse the second voice information; The parsing module is further configured to generate and play the guiding information according to the second voice information if the parsing of the second voice information fails; The parsing module is also used to determine whether there is a third voice information in the first database, wherein the similarity between the third voice information and the second voice information is greater than or equal to a preset threshold; if the third voice information exists in the first database, the guide information is generated according to the third voice information; if the third voice information does not exist in the first database, it is determined whether the second voice information is small talk information; if the second voice information is small talk information, third feedback information is played, wherein the third feedback information is small talk words; if the second voice information is not small talk information, fourth feedback information is played, wherein the fourth feedback information is used to indicate that voice understanding fails.

6. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the speech processing method according to any one of claims 1 to 4 is implemented.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the speech processing method according to any one of claims 1 to 4 is implemented.

Images