Business process identification methods, devices, equipment and storage media

Abstract

This invention discloses a business process identification method, apparatus, device, and storage medium. The business process identification method includes: acquiring an audio file to be identified; performing speech recognition on the audio file to obtain an audio recognition result; executing a preset business process chain to call a process state matcher to match the audio recognition result, thereby obtaining a business process identification result. This invention can improve the accuracy of business process identification results.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a business process identification method, apparatus, device, and storage medium. Background Technology

[0002] With the advancement and deepening application of AI technology, the core effects, product maturity, and business application models of machine learning-based vision, image, speech, and language analysis technologies have reached a practical level. Currently, AI technology is widely used in industries such as telecommunications operators, banking, insurance, securities, power, and transportation, becoming an essential technical support for scenarios such as on-site business compliance and remote customer service, and is also a focus of attention for major insurance companies in China.

[0003] Intelligent Dual-Recording System (IDRS) addresses compliance requirements for dual-recording (audio and video) scenarios in the financial sectors such as insurance and banking. Utilizing AI technology, it detects various key actions, voice recordings, and documents during the sales process, as well as facial tracking of key personnel, enabling intelligent monitoring of the sales process. The App-based detection capabilities provided by IDRS allow most sales compliance checks to be completed on-site, enabling real-time problem detection, alerts, and corrections, significantly improving sales friendliness and first-time pass rates.

[0004] When performing business process recognition on audio files obtained from intelligent dual-recording quality inspection, the current practice is generally to first perform speech recognition on the audio files to obtain audio recognition text, and then use NLP (Natural Language Processing) intent recognition algorithms to perform rule matching on the audio recognition text. However, NLP algorithms are not true artificial intelligence, their recognition effect is generally average, and they are not flexible enough, resulting in poor accuracy of business process recognition results. Summary of the Invention

[0005] The main objective of this invention is to provide a business process identification method, apparatus, device, and storage medium, which aims to solve the problem of poor accuracy of existing business process identification results.

[0006] To achieve the above objectives, the present invention provides a business process identification method, the business process identification method comprising:

[0007] Obtain the audio file to be recognized;

[0008] The audio file to be identified is subjected to speech recognition to obtain the audio recognition result;

[0009] Execute the preset business process chain to call the process state matcher to match the audio recognition result and obtain the business process recognition result.

[0010] Preferably, before the step of performing speech recognition on the audio file to be recognized to obtain the audio recognition result, the business process recognition method further includes:

[0011] The audio file to be identified is subjected to VAD detection to obtain the detection results;

[0012] Based on the detection results, determine whether the audio file to be identified is blank audio;

[0013] If the audio file to be identified is not blank, then the following steps are performed: perform speech recognition on the audio file to be identified to obtain the audio recognition result.

[0014] Preferably, before the step of performing speech recognition on the audio file to be recognized to obtain the audio recognition result, the business process recognition method further includes:

[0015] Based on the detection results, determine whether there are silent segments at the beginning and / or end of the audio file to be identified;

[0016] If there are silent segments at the beginning and / or end of the audio file to be identified, the silent time period is determined based on the detection results.

[0017] The target audio file is obtained by truncating the audio file to be identified based on the silent period.

[0018] The step of performing speech recognition on the audio file to be recognized to obtain the audio recognition result includes:

[0019] The target audio file is subjected to speech recognition to obtain the audio recognition result.

[0020] Preferably, before the step of executing a preset business process chain to call a process state matcher to match the audio recognition result and obtain the business process recognition result, the business process recognition method further includes:

[0021] Obtain business process information, wherein the business process information includes business process node number, business process script, keywords and business process sequence;

[0022] The keywords are constructed into a state tree of a finite state machine based on the finite state machine algorithm, and the process state matcher corresponding to each business process node is obtained based on the state tree.

[0023] The business process script and the process status matcher are assembled according to the business process node number and the business process sequence to obtain the preset business process chain.

[0024] Preferably, the step of executing a preset business process chain to call a process state matcher to match the audio recognition result and obtain the business process recognition result includes:

[0025] Execute a preset business process chain to sequentially call the corresponding process state matcher according to the business process order, and match the keywords in the state tree of the called process state matcher with the audio recognition result to obtain the matching result;

[0026] Based on the matching results, the time points corresponding to each business process node in the audio file to be identified are determined, wherein the business process identification results include the matching results and the time points corresponding to each business process node in the audio file to be identified.

[0027] Preferably, the business process identification method further includes:

[0028] Obtain the time point corresponding to each business process node in the business process recognition result in the audio file to be recognized.

[0029] Based on the corresponding time points of each business process node in the audio file to be identified, the audio file to be identified is segmented to obtain key audio segments;

[0030] The audio file to be identified is subjected to speech recognition to obtain key business information.

[0031] Furthermore, to achieve the above objectives, the present invention also provides a business process identification device, the business process identification device comprising:

[0032] The file acquisition module is used to acquire the audio file to be recognized;

[0033] The speech recognition module is used to perform speech recognition on the audio file to be recognized and obtain the audio recognition result;

[0034] The process recognition module is used to execute a preset business process chain to call the process state matcher to match the audio recognition result and obtain the business process recognition result.

[0035] In addition, to achieve the above objectives, the present invention also provides a business process identification device, the business process identification device comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the steps of the business process identification method as described above.

[0036] In addition, to achieve the above objectives, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the business process identification method described above.

[0037] In addition, to achieve the above objectives, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the business process identification method described above.

[0038] This invention provides a method, apparatus, device, storage medium, and product for business process identification. First, an audio file to be identified is acquired. Then, speech recognition is performed on the audio file to obtain the audio recognition result. Next, a preset business process chain is executed to call a process state matcher to match the audio recognition result, thus obtaining the business process identification result. The preset business process chain is constructed using the chain of responsibility pattern, and the process state matcher is constructed based on a finite state machine. This invention combines the chain of responsibility pattern with a finite state machine to complete the matching and identification of business processes. Compared to using NLP intent recognition algorithms, this significantly improves the accuracy and precision of business process identification. Furthermore, when using NLP algorithms for identification, different business processes require the wording of the business process to be organized beforehand and given to algorithm engineers for NLP rule configuration, followed by the generation of a new NLP algorithm deployment package before rule recognition can be performed on the new process. This process is complex and inefficient. Therefore, this invention, compared to the above method, eliminates the need for manual configuration of relevant rules, greatly simplifying the process complexity and improving identification efficiency. Attached Figure Description

[0039] Figure 1 This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention;

[0040] Figure 2 This is a flowchart illustrating the first embodiment of the business process identification method of the present invention;

[0041] Figure 3 This is a schematic diagram of the functional modules of the first embodiment of the business process identification device of the present invention.

[0042] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0043] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0044] Reference Figure 1 , Figure 1This is a schematic diagram of the device structure of the hardware operating environment involved in the embodiments of the present invention.

[0045] The business process identification device in this embodiment of the invention can be a server, or a terminal device such as a PC (Personal Computer), tablet computer, or portable computer.

[0046] like Figure 1 As shown, the business process identification device may include: a processor 1001, such as a CPU; a communication bus 1002; a user interface 1003; a network interface 1004; and a memory 1005. The communication bus 1002 is used to establish communication between these components. The user interface 1003 may include a display screen or an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface). The memory 1005 may be high-speed RAM or non-volatile memory, such as a disk drive. Optionally, the memory 1005 may also be a storage device independent of the aforementioned processor 1001.

[0047] Those skilled in the art will understand that Figure 1 The business process identification device structure shown does not constitute a limitation on the business process identification device. It may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0048] like Figure 1 As shown, the memory 1005, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and computer programs.

[0049] exist Figure 1 In the terminal shown, network interface 1004 is mainly used to connect to the backend server and communicate with it; user interface 1003 is mainly used to connect to the client and communicate with it; and processor 1001 can be used to call computer programs stored in memory 1005 and perform the following operations:

[0050] Obtain the audio file to be recognized;

[0051] The audio file to be identified is subjected to speech recognition to obtain the audio recognition result;

[0052] Execute the preset business process chain to call the process state matcher to match the audio recognition result and obtain the business process recognition result.

[0053] Furthermore, the processor 1001 can call a computer program stored in the memory 1005 and also perform the following operations:

[0054] The audio file to be identified is subjected to VAD detection to obtain the detection results;

[0055] Based on the detection results, determine whether the audio file to be identified is blank audio;

[0056] If the audio file to be identified is not blank, then speech recognition is performed on the audio file to be identified to obtain the audio recognition result.

[0057] Furthermore, the processor 1001 can call a computer program stored in the memory 1005 and also perform the following operations:

[0058] Based on the detection results, determine whether there are silent segments at the beginning and / or end of the audio file to be identified;

[0059] If there are silent segments at the beginning and / or end of the audio file to be identified, the silent time period is determined based on the detection results.

[0060] The target audio file is obtained by truncating the audio file to be identified based on the silent period.

[0061] The target audio file is subjected to speech recognition to obtain the audio recognition result.

[0062] Furthermore, the processor 1001 can call a computer program stored in the memory 1005 and also perform the following operations:

[0063] Obtain business process information, wherein the business process information includes business process node number, business process script, keywords and business process sequence;

[0064] The keywords are constructed into a state tree of a finite state machine based on the finite state machine algorithm, and the process state matcher corresponding to each business process node is obtained based on the state tree.

[0065] The business process script and the process status matcher are assembled according to the business process node number and the business process sequence to obtain the preset business process chain.

[0066] Furthermore, the processor 1001 can call a computer program stored in the memory 1005 and also perform the following operations:

[0067] Execute a preset business process chain to sequentially call the corresponding process state matcher according to the business process order, and match the keywords in the state tree of the called process state matcher with the audio recognition result to obtain the matching result;

[0068] Based on the matching results, the time points corresponding to each business process node in the audio file to be identified are determined, wherein the business process identification results include the matching results and the time points corresponding to each business process node in the audio file to be identified.

[0069] Furthermore, the processor 1001 can call a computer program stored in the memory 1005 and also perform the following operations:

[0070] Obtain the time point corresponding to each business process node in the business process recognition result in the audio file to be recognized.

[0071] Based on the corresponding time points of each business process node in the audio file to be identified, the audio file to be identified is segmented to obtain key audio segments;

[0072] The audio file to be identified is subjected to speech recognition to obtain key business information.

[0073] Based on the above hardware structure, various embodiments of the business process identification method of the present invention are proposed.

[0074] This invention provides a business process identification method.

[0075] Reference Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the business process identification method of the present invention.

[0076] In this embodiment, the business process identification method includes:

[0077] Step S10: Obtain the audio file to be recognized;

[0078] The business process identification method in this embodiment is implemented by a business process identification device, which can be a server, or a terminal device such as a PC (Personal Computer), tablet computer, or portable computer.

[0079] In this embodiment, the audio file to be identified is first obtained. The audio file to be identified is a pre-recorded audio file, and the recording scenarios include, but are not limited to: intelligent dual-recording quality inspection scenarios (i.e., audio and video recording scenarios), product sales recording scenarios (i.e., simple audio recording scenarios), etc.

[0080] It should be noted that in the case of intelligent dual-recording quality inspection, video and audio files will be obtained. In this case, the audio part can be extracted and segmented first to obtain the audio file, which can be used as the audio file to be identified in this application.

[0081] Step S20: Perform speech recognition on the audio file to be recognized to obtain the audio recognition result;

[0082] After obtaining the audio file to be recognized, speech recognition is performed on the audio file to obtain the audio recognition result. During speech recognition, the ASR (Automatic Speech Recognition) algorithm can be used, and the corresponding interface can be called for speech recognition. Alternatively, in specific implementations, the audio file to be recognized can be sent to a professional platform for speech recognition, and then the audio recognition result returned by the professional platform can be received.

[0083] The audio recognition results include, but are not limited to: channel_id (the ID of the audio track to which the sentence belongs), begin_time (the start time offset of the sentence in milliseconds), end_time (the end time offset of the sentence in milliseconds), text (the recognized text result of the sentence), emotion_value (emotional energy value 1-10, the higher the value, the stronger the emotion), silence_duration (the silence duration between this sentence and the previous sentence in seconds), and speech_rate (the average speech rate of this sentence in words per minute).

[0084] Step S30: Execute the preset business process chain to call the process state matcher to match the audio recognition result and obtain the business process recognition result.

[0085] After obtaining the audio recognition result through speech recognition, a preset business process chain is executed. This preset business process chain is pre-assembled, and the specific assembly process can be referred to the fourth embodiment below, which will not be elaborated here.

[0086] This preset business process chain contains the script text for each business process node and a process status matcher. When executing this preset business process chain, each business process node can be entered sequentially according to the business process order. After entering the corresponding business process node, the corresponding process status matcher can be called, and the keywords in the state tree of the called process status matcher are matched with the audio recognition results. When each business process node in the business process chain has been matched, the matching result can be obtained. Then, based on the matching result, the corresponding time point of each business process node in the audio file to be recognized can be determined. The business process recognition result includes the matching result and the corresponding time point of each business process node in the audio file to be recognized.

[0087] In addition, it should be noted that if there are multiple preset business process chains, after obtaining the audio file to be recognized, it is necessary to first determine the target business process chain from the multiple preset business process chains based on the audio file to be recognized. Then, after obtaining the audio recognition result from the speech recognition, the target business process chain is executed to call the process state matcher to match the audio recognition result and obtain the business process recognition result.

[0088] This invention provides a business process recognition method. First, an audio file to be recognized is acquired. Then, speech recognition is performed on the audio file to obtain the audio recognition result. Next, a preset business process chain is executed to call a process state matcher to match the audio recognition result, thus obtaining the business process recognition result. The preset business process chain is constructed using the chain of responsibility pattern, and the process state matcher is constructed based on a finite state machine. This invention combines the chain of responsibility pattern with a finite state machine to complete the matching and recognition of business processes. Compared to using NLP intent recognition algorithms, this significantly improves the accuracy and precision of business process recognition. Furthermore, when using NLP algorithms for recognition, different business processes require the wording of the business process to be organized beforehand and given to algorithm engineers for NLP rule configuration. A new NLP algorithm deployment package is then generated before rule recognition can be performed on the new process, making the process complex and inefficient. Therefore, this invention, compared to the above method, eliminates the need for manual rule configuration, greatly simplifying the process complexity and improving recognition efficiency.

[0089] Furthermore, based on the first embodiment described above, a second embodiment of the business process identification method of the present invention is proposed.

[0090] In this embodiment, prior to step S20, the business process identification method further includes:

[0091] Step A: Perform VAD detection on the audio file to be identified and obtain the detection result;

[0092] In this embodiment, after obtaining the audio file to be recognized, before calling the algorithm to perform speech recognition, the audio file to be recognized is first subjected to VAD (Voice Activity Detection, also known as speech endpoint detection or speech boundary detection) detection to obtain the detection results in order to identify silent segments from the sound signal stream.

[0093] Step B: Determine whether the audio file to be identified is blank audio based on the detection results;

[0094] If the audio file to be identified is not blank, then step S20 is executed: speech recognition is performed on the audio file to be identified to obtain the audio recognition result.

[0095] After obtaining the VAD detection results, the system determines whether the audio file to be identified is blank audio based on the detection results. Specifically, it checks whether the total duration of the silent segments in the audio file to be identified is greater than or equal to the product of the total duration and a preset ratio (which can be set according to actual needs). If the total duration of the silent segments is greater than or equal to the product value, the audio file to be identified is determined to be blank audio; if the total duration of the silent segments is less than the product value, the audio file to be identified is determined not to be blank audio.

[0096] Furthermore, if the audio file to be identified is not blank, the next step is to perform speech recognition on the audio file to obtain the audio recognition result. The specific execution process can be referred to the first embodiment described above, and will not be repeated here.

[0097] Furthermore, if the audio file to be identified is blank, the subsequent steps are stopped, and an error message is generated to indicate that the audio file to be identified is blank and does not need to be detected, so as to avoid wasting algorithm resources.

[0098] In this embodiment, by performing VAD detection on the audio file to be recognized, subsequent detection can be skipped when blank audio is detected. Subsequent detection will only be performed when the audio file to be recognized is not blank audio. In this way, the waste of algorithm resources can be reduced while the cost incurred by enterprises due to calling speech recognition algorithms can be reduced.

[0099] Furthermore, based on the second embodiment described above, a third embodiment of the business process identification method of the present invention is proposed.

[0100] In this embodiment, prior to step S20, the business process identification method further includes:

[0101] Step C: Determine whether there are silent segments at the beginning and / or end of the audio file to be identified based on the detection results;

[0102] In this embodiment, after obtaining the VAD detection result and detecting blank audio, if the audio file to be identified is not blank audio, it can be further determined whether there are silent segments at the beginning and / or end of the audio file to be identified based on the detection result.

[0103] Step D: If there are silent segments at the beginning and / or end of the audio file to be identified, the silent time period is determined based on the detection results.

[0104] Step E: Extract the audio file to be identified based on the silent period to obtain the target audio file;

[0105] At this point, step S20 may include:

[0106] The target audio file is subjected to speech recognition to obtain the audio recognition result.

[0107] If there are silent segments at the beginning and / or end of the audio file to be identified, the silent time period is determined based on the detection results, that is, the time period corresponding to the silent segments at the beginning and / or end. Then, the audio file to be identified is truncated according to the silent time period, that is, the silent segments at the beginning and / or end are removed, and the non-silent segments in the middle are retained to obtain the target audio file. Then, speech recognition is performed on the target audio file to obtain the audio recognition result, and subsequent steps are further executed. The specific execution process can be referred to the first embodiment above, and will not be repeated here.

[0108] By using the above method, the presence of silent segments at the beginning and / or end of the audio file to be recognized is detected. If such segments are found, they are segmented out, thereby further avoiding the waste of some algorithm resources and the waste of enterprise costs in calling speech recognition algorithms, thus saving enterprise costs.

[0109] Furthermore, based on the first to third embodiments described above, a fourth embodiment of the business process identification method of the present invention is proposed.

[0110] In this embodiment, prior to step S30, the business process identification method further includes:

[0111] Step F: Obtain business process information, wherein the business process information includes business process node number, business process script, keywords, and business process sequence;

[0112] Step G: Construct the keyword into a state tree of a finite state machine according to the finite state machine algorithm, and obtain the process state matcher corresponding to each business process node based on the state tree;

[0113] Step H: Assemble the business process script and the process status matcher according to the business process node number and the business process sequence to obtain the preset business process chain.

[0114] In this embodiment, business process information is acquired. This information includes the business process node number (nodeId), business process speech (Speech), keywords (Keyword), and business process sequence (detailSort). The business process information is typically expressed as JSON-formatted message data. The following example illustrates this process using an intelligent dual-recording quality inspection service in the sales process of a financial product.

[0115] Business process information can be:

[0116]

[0117]

[0118] After obtaining the business process information, the keywords can be used to construct a state tree of the finite state machine using the finite state machine algorithm. Based on the state tree, a process state matcher corresponding to each business process node can be obtained. A finite state machine (DFA) refers to a process that transitions from one state to another through a series of events (state->event->state). It is a tool used for object behavior modeling, primarily describing the sequence of states an object experiences during its lifecycle and how it responds to various external events.

[0119] After constructing the process status matcher, the business process scripts and process status matchers are assembled according to the chain of responsibility pattern based on the business process node number and business process sequence to obtain a preset business process chain. Finally, the preset business process chain contains the script text and process status matcher of each business process node.

[0120] It's important to note that the Chain of Responsibility pattern aims to avoid coupling the request sender with multiple request handlers. It links all request handlers into a chain by having each object remember a reference to the next object. When a request occurs, it is passed along this chain until an object handles it. In the Chain of Responsibility pattern, the client only needs to send the request to the chain; it doesn't need to worry about the processing details or the transmission process, as the request is automatically passed. Therefore, the Chain of Responsibility decouples the request sender from the request handler. The Chain of Responsibility pattern is a behavioral object design pattern with the following main advantages: ① Reduced coupling between objects. This pattern allows an object to be unaware of which object handles its request or the chain structure, and the sender and receiver do not need explicit information about each other. This enhances system scalability; ② New request handling classes can be added as needed, satisfying the Open / Closed Principle. This enhances the flexibility of assigning responsibilities to objects. When the workflow changes, the members in the chain can be dynamically changed or their order adjusted, and responsibilities can be dynamically added or removed; ③ The Chain of Responsibility simplifies the connections between objects. Each object only needs to maintain a reference to its successor, and does not need to maintain references to all other handlers, which avoids the use of numerous if or ifelse statements; ④ Responsibility sharing. Each class only needs to handle its own work, and the work that should not be handled is passed to the next object, which clarifies the scope of responsibility of each class and conforms to the single responsibility principle of classes.

[0121] Further, step S30 may include:

[0122] Step a31: Execute the preset business process chain to sequentially call the corresponding process state matcher according to the business process order, and match the keywords in the state tree of the called process state matcher with the audio recognition result to obtain the matching result;

[0123] Step a32: Determine the time point corresponding to each business process node in the audio file to be identified based on the matching result, wherein the business process identification result includes the matching result and the time point corresponding to each business process node in the audio file to be identified.

[0124] In this embodiment, when performing business process recognition based on the audio recognition results obtained from speech recognition, the specific recognition process is as follows: A preset business process chain is executed, thereby sequentially entering each business process node according to the business process order. After entering the corresponding business process node, the corresponding process state matcher can be called, and the keywords in the state tree of the called process state matcher are matched with the audio recognition results. When each business process node in the business process chain has been matched, a matching result is obtained. Furthermore, the time point corresponding to each business process node in the audio file to be recognized can be determined based on the matching result. The business process recognition result includes the matching result and the time point corresponding to each business process node in the audio file to be recognized.

[0125] For example, let's take the business process information in the example above as an illustration.

[0126] Assuming that after performing speech recognition on the audio file corresponding to the above business process information, the customer obtains the following audio recognition result:

[0127]

[0128]

[0129]

[0130] Finally, after further identifying the business process of the audio recognition file through a preset business process chain, the business process recognition results are as follows:

[0131]

[0132]

[0133]

[0134] In this embodiment, a business process chain is dynamically constructed using the chain of responsibility pattern. Then, a finite state machine algorithm is used to build a state tree of keywords to be matched, performing matching one by one until all nodes in the business process chain are matched, ultimately obtaining the matching result and thus the corresponding time point of the business process dialogue in the audio file to be recognized. By combining the chain of responsibility pattern with a finite state machine to complete the matching and recognition of business processes, compared to using NLP intent recognition algorithms, the accuracy and precision of business process recognition can be greatly improved. Furthermore, the elimination of the need for manual configuration of relevant rules significantly simplifies the process complexity and improves recognition efficiency.

[0135] Furthermore, based on the first to third embodiments described above, a fifth embodiment of the business process identification method of the present invention is proposed.

[0136] In this embodiment, after step S30, the business process identification method further includes:

[0137] Step 1: Obtain the time point corresponding to each business process node in the business process recognition result in the audio file to be recognized;

[0138] In this embodiment, after obtaining the business process identification result, the corresponding time point of each business process node in the business process identification result in the audio file to be identified can be obtained.

[0139] Step J: Extract key audio segments from the audio file to be identified based on the time points corresponding to the business process nodes in the audio file to be identified.

[0140] Step K involves performing speech recognition on the audio file to be recognized to obtain key business information.

[0141] Then, based on the corresponding time points of each business process node in the audio file to be identified, the key audio segments can be extracted to obtain the customer response audio segments corresponding to each business process node. Then, the ASR algorithm can be used to perform speech recognition on the audio file to be identified to obtain key business information, namely the customer response answer.

[0142] By using the methods described above, customer responses (i.e., key business information) in the business process can be accurately located, thereby enabling the rapid acquisition of customer responses.

[0143] The present invention also provides a business process identification device.

[0144] Reference Figure 3 , Figure 3 This is a schematic diagram of the functional modules of the first embodiment of the business process identification device of the present invention.

[0145] like Figure 3 As shown, the business process identification device includes:

[0146] File acquisition module 10 is used to acquire the audio file to be recognized;

[0147] The speech recognition module 20 is used to perform speech recognition on the audio file to be recognized and obtain the audio recognition result;

[0148] The process recognition module 30 is used to execute a preset business process chain to call the process state matcher to match the audio recognition result and obtain the business process recognition result.

[0149] Furthermore, the business process identification device also includes:

[0150] The VAD detection module is used to perform VAD detection on the audio file to be identified and obtain the detection result;

[0151] The first judgment module is used to determine whether the audio file to be identified is blank audio based on the detection result;

[0152] The speech recognition module 20 is specifically used to perform speech recognition on the audio file to be recognized if the audio file to be recognized is not blank audio, and obtain the audio recognition result.

[0153] Furthermore, the business process identification device also includes:

[0154] The second judgment module is used to determine whether there are silent segments at the beginning and / or end of the audio file to be identified based on the detection results.

[0155] The time period determination module is used to determine the silent time period based on the detection result if there is a silent segment at the beginning and / or end of the audio file to be identified.

[0156] The first segmentation module is used to segment the audio file to be identified according to the silent period to obtain the target audio file;

[0157] The speech recognition module 20 is also specifically used for:

[0158] The target audio file is subjected to speech recognition to obtain the audio recognition result.

[0159] Furthermore, the business process identification device also includes:

[0160] The information acquisition module is used to acquire business process information, wherein the business process information includes business process node number, business process script, keywords and business process sequence;

[0161] The construction module is used to construct the keywords into a state tree of a finite state machine according to the finite state machine algorithm, and obtain the process state matcher corresponding to each business process node based on the state tree.

[0162] An assembly module is used to assemble the business process script and the process status matcher according to the business process node number and the business process sequence to obtain the preset business process chain.

[0163] Furthermore, the process identification module includes:

[0164] The matching unit is used to execute a preset business process chain, so as to sequentially call the corresponding process state matcher according to the business process order, and match the keywords in the state tree of the called process state matcher with the audio recognition result to obtain the matching result;

[0165] The determining unit is used to determine the time point corresponding to each business process node in the audio file to be identified based on the matching result, wherein the business process identification result includes the matching result and the time point corresponding to each business process node in the audio file to be identified.

[0166] Furthermore, the business process identification device also includes:

[0167] The time point acquisition module is used to acquire the time point corresponding to each business process node in the business process recognition result in the audio file to be recognized.

[0168] The second extraction module is used to extract key audio segments from the audio file to be identified based on the time points corresponding to the business process nodes in the audio file to be identified.

[0169] The speech recognition module 20 is also used to perform speech recognition on the audio file to be recognized to obtain key business information.

[0170] The functions of each module in the above-mentioned business process identification device correspond to the steps in the above-mentioned business process identification method embodiment, and their functions and implementation processes will not be described in detail here.

[0171] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the steps of the business process identification method as described in any of the above embodiments.

[0172] The specific embodiments of the computer-readable storage medium of the present invention are basically the same as the embodiments of the above-described business process identification method, and will not be described in detail here.

[0173] The present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the business process identification method as described in any of the above claims.

[0174] The specific embodiments of the computer program product of the present invention are basically the same as the embodiments of the above-described business process identification method, and will not be described in detail here.

[0175] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or system that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or system. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or system that includes that element.

[0176] The sequence numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0177] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0178] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. A business process identification method, characterized by The business process identification method includes: Obtain the audio file to be recognized; The audio file to be identified is subjected to speech recognition to obtain the audio recognition result; A preset business process chain is executed to call the process status matcher to match the audio recognition result, thereby obtaining the business process recognition result; the preset business process chain includes at least the speech text of each business process node and the process status matcher. Prior to the step of executing a preset business process chain to call a process state matcher to match the audio recognition result and obtain the business process recognition result, the business process recognition method further includes: Obtain business process information, wherein the business process information includes business process node number, business process script, keywords and business process sequence; The keywords are constructed into a state tree of a finite state machine based on the finite state machine algorithm, and the process state matcher corresponding to each business process node is obtained based on the state tree. The business process script and the process status matcher are assembled according to the business process node number and the business process sequence to obtain the preset business process chain; The step of executing a preset business process chain to call a process state matcher to match the audio recognition result and obtain the business process recognition result includes: Execute a preset business process chain to sequentially call the corresponding process state matcher according to the business process order, and match the keywords in the state tree of the called process state matcher with the audio recognition result to obtain the matching result; Based on the matching results, the time points corresponding to each business process node in the audio file to be identified are determined, wherein the business process identification results include the matching results and the time points corresponding to each business process node in the audio file to be identified.

2. The business process identification method of claim 1, wherein, Before the step of performing speech recognition on the audio file to be recognized to obtain the audio recognition result, the business process recognition method further includes: The audio file to be identified is subjected to VAD detection to obtain the detection results; Based on the detection results, determine whether the audio file to be identified is blank audio; If the audio file to be identified is not blank, then the following step is performed: perform speech recognition on the audio file to be identified to obtain the audio recognition result.

3. The business process identification method of claim 2, wherein, Before the step of performing speech recognition on the audio file to be recognized to obtain the audio recognition result, the business process recognition method further includes: Based on the detection results, determine whether there are silent segments at the beginning and / or end of the audio file to be identified; If there are silent segments at the beginning and / or end of the audio file to be identified, the silent time period is determined based on the detection results. The audio file to be identified is truncated according to the silent period to obtain the target audio file; the step of performing speech recognition on the audio file to be identified to obtain the audio recognition result includes: The audio file to be identified is subjected to speech recognition to obtain the audio recognition result.

4. The business process identification method according to any one of claims 1 to 3, characterized by, The business process identification method also includes: Obtain the time point corresponding to each business process node in the business process recognition result in the audio file to be recognized. Based on the corresponding time points of each business process node in the audio file to be identified, the audio file to be identified is segmented to obtain key audio segments; The audio file to be identified is subjected to speech recognition to obtain key business information.

5. A business process recognition apparatus characterized by comprising: The business process identification device includes: The file acquisition module is used to acquire the audio file to be recognized; The speech recognition module is used to perform speech recognition on the audio file to be recognized and obtain the audio recognition result; The process recognition module is used to execute a preset business process chain to call the process status matcher to match the audio recognition result and obtain the business process recognition result; the preset business process chain includes at least the speech text of each business process node and the process status matcher. Prior to the step of executing a preset business process chain to call a process state matcher to match the audio recognition result and obtain the business process recognition result, the process recognition module further includes: The information acquisition module is used to acquire business process information, wherein the business process information includes business process node number, business process script, keywords and business process sequence; The construction module is used to construct the keywords into a state tree of a finite state machine according to the finite state machine algorithm, and obtain the process state matcher corresponding to each business process node based on the state tree. An assembly module is used to assemble the business process script and the process status matcher according to the business process node number and the business process sequence to obtain the preset business process chain; The step of executing a preset business process chain to call a process state matcher to match the audio recognition result and obtain the business process recognition result includes: Execute a preset business process chain to sequentially call the corresponding process state matcher according to the business process order, and match the keywords in the state tree of the called process state matcher with the audio recognition result to obtain the matching result; Based on the matching results, the time points corresponding to each business process node in the audio file to be identified are determined, wherein the business process identification results include the matching results and the time points corresponding to each business process node in the audio file to be identified.

6. A business process recognition apparatus characterized by comprising: The business process identification device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the steps of the business process identification method as described in any one of claims 1 to 4.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the business process identification method as described in any one of claims 1 to 4.

Images