Medical voice service quality inspection method, device, equipment and medium

Abstract

The invention relates to the field of intelligent medical treatment, and discloses a medical voice service quality inspection method, device and equipment and a medium, and the method comprises the steps: monitoring a medical service voice process to obtain real-time voice data, and recognizing first voice data of a medical consultation person and second voice data of a medical customer service person; calling a voice quality inspection template according to the first voice data to determine a target intonation and a target keyword sentence; performing quality inspection on the second voice data according to the target intonation and the target keyword sentence to obtain real-time reminding information so as to remind medical customer service personnel; when the scoring condition is met, quality inspection scoring is carried out according to the first voice data and the second voice data, the voice quality inspection template is called based on the first voice data, and the second voice data is subjected to quality inspection according to the voice quality inspection template so as to remind medical customer service personnel in real time; and medical customer service personnel can be effectively assisted to provide high-quality and standardized medical voice service for medical consultation personnel.

Description

Technical Field

[0001] The present application relates to the field of intelligent medical care, and in particular to a medical voice service quality inspection method, device, equipment and medium. Background Art

[0002] Smart healthcare is a recently emerging medical term that uses the most advanced Internet of Things technology to achieve interaction between patients and medical staff, medical institutions, and medical equipment, gradually achieving informatization. Among them, an important application of smart healthcare is that users can consult medical staff through telephone or other voice communication, thereby realizing online consultation.

[0003] However, in the process of voice communication between medical staff and users, there is a lack of a voice quality inspection system that is closely integrated with medical services and supporting effective intervention measures, which is not conducive to improving the standardization of services provided by relevant medical staff and helping patients build trust in online consultations. Summary of the invention

[0004] The embodiments of the present invention provide a medical voice service quality inspection method, device, equipment and medium, which are intended to perform quality inspection on medical voice services to provide real-time reminders to medical customer service personnel, and help provide high-quality and standardized medical voice services.

[0005] In a first aspect, an embodiment of the present invention provides a medical voice service quality inspection method, comprising:

[0006] When receiving a medical voice service quality inspection instruction, determining the medical service voice process corresponding to the medical voice service quality inspection instruction, and monitoring the medical service voice process to obtain real-time voice data;

[0007] Recognize the real-time voice data to obtain first voice data corresponding to the medical consultant and second voice data corresponding to the medical customer service staff;

[0008] Retrieving a matching voice quality inspection template according to the first voice data;

[0009] Parse the speech quality inspection template to determine the target intonation and target key words and phrases corresponding to the speech quality inspection template;

[0010] Performing a synchronous quality inspection operation on the second voice data according to the target intonation and the target key words and phrases to generate real-time reminder information, and sending the real-time reminder information to the operation terminal of the medical customer service personnel;

[0011] When the preset scoring conditions are met, a quality inspection score of the medical service process is generated according to the first voice data and the second voice data.

[0012] In some embodiments, recognizing real-time voice data to obtain first voice data corresponding to a medical consultant and second voice data corresponding to a medical customer service staff includes:

[0013] Inputting the real-time speech data into a preset speech conversion model for text conversion to generate a plurality of segment text information, wherein each segment text information is respectively associated with a different label time;

[0014] The preset speech feature recognition model is used to identify the speech features corresponding to each label time in the real-time speech data;

[0015] The first voice data and the second voice data are determined from the real-time voice data according to the voice text information and the voice feature.

[0016] In some implementations, the speech features corresponding to each label time in the real-time speech data are identified using a preset speech feature recognition model, including:

[0017] Extract corresponding sentence segment voice data from real-time voice data according to the label time;

[0018] The sentence speech data is input into the speech feature recognition model for speech feature recognition to generate corresponding speech features.

[0019] In some implementations, determining the first voice data and the second voice data from the real-time voice data according to the voice text information and the voice feature includes:

[0020] Determine the probability of a first sound source corresponding to the label time based on the voice text information;

[0021] Determine the probability of a second sound source corresponding to the label time based on the voice feature;

[0022] The sound source corresponding to the label time is determined to be a medical consultant or a medical customer service staff according to the first sound source probability and the second sound source probability, so as to determine the first voice data and the second voice data from the real-time voice data.

[0023] In some implementations, the real-time reminder information includes at least one of intonation reminder information and key word reminder information, and performing a synchronous quality inspection operation on the second voice data according to the target intonation and the target key word includes:

[0024] Recognize a first real-time intonation of the first voice data and a second real-time intonation of the second voice data; generate intonation reminder information according to the target intonation, the first real-time intonation and the second real-time intonation;

[0025] and / or,

[0026] Determine the frequency information of the target key words and phrases in the target key words and phrases, and generate key word and phrase reminder information according to the frequency information.

[0027] In some implementations, the target keyword phrase includes at least one target sub-word phrase; generating keyword phrase reminder information according to the frequency information includes:

[0028] Determine the occurrence frequency corresponding to the target sub-word sentence according to the frequency information;

[0029] The occurrence frequency is compared with the target frequency to generate sub-reminder information corresponding to the target sub-word and sentence, and the sub-reminder information is integrated to generate the word and sentence reminder information, wherein the target frequency is associated with the target sub-word and sentence.

[0030] In some embodiments, generating a quality inspection score of the medical service process according to the first voice data and the second voice data includes:

[0031] Performing semantic recognition on the first voice data to obtain consultation content information corresponding to the first voice data;

[0032] Extracting keywords from the second voice data to obtain answer keywords;

[0033] The accuracy of the medical customer service staff's responses is determined based on the consultation content information and response keywords, thereby generating a quality inspection score.

[0034] In a second aspect, an embodiment of the present invention further provides a medical voice service quality inspection device, comprising:

[0035] A voice monitoring module is used to determine the medical service voice process corresponding to the medical voice service quality inspection instruction when receiving the medical voice service quality inspection instruction, and monitor the medical service voice process to obtain real-time voice data;

[0036] A sound source recognition module, used to recognize real-time voice data to obtain first voice data corresponding to the medical consultant and second voice data corresponding to the medical customer service staff;

[0037] A template retrieval module, used to retrieve a matching voice quality inspection template according to the first voice data;

[0038] A template parsing module, used to parse the speech quality inspection template to determine the target intonation and target key words and phrases corresponding to the speech quality inspection template;

[0039] A first quality inspection module is used to perform a synchronous quality inspection operation on the second voice data according to the target intonation and the target key words and phrases to generate real-time reminder information, and send the real-time reminder information to the operation terminal of the medical customer service personnel;

[0040] The second quality inspection module is used to generate a quality inspection score for the medical service process according to the first voice data and the second voice data when a preset scoring condition is met.

[0041] In a third aspect, an embodiment of the present invention further provides a computer device, the computer device comprising a memory and a processor;

[0042] Memory for storing computer programs;

[0043] A processor is used to execute a computer program and implement any medical voice service quality inspection method provided in the embodiments of this specification when executing the computer program.

[0044] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the processor implements any medical voice service quality inspection method provided in the embodiments of this specification.

[0045] The embodiment of the present invention provides a medical voice service quality inspection method, apparatus, computer equipment and computer-readable storage medium, wherein the medical voice service quality inspection method comprises: when a medical voice service quality inspection instruction is received, determining the medical service voice process corresponding to the medical voice service quality inspection instruction, and monitoring the medical service voice process to obtain real-time voice data; identifying the real-time voice data to obtain first voice data corresponding to the medical consultant and second voice data corresponding to the medical customer service personnel; retrieving a matching voice quality inspection template according to the first voice data; parsing the voice quality inspection template to determine the target intonation and target key words and phrases corresponding to the voice quality inspection template; and retrieving the matching voice quality inspection template according to the target intonation and target key words and phrases. The second voice data is synchronously quality inspected by the word and sentence to generate real-time reminder information, and the real-time reminder information is sent to the operation terminal of the medical customer service personnel; when the preset scoring conditions are met, a quality inspection score of the medical service process is generated according to the first voice data and the second voice data, by monitoring the medical service voice process, and retrieving the matching voice quality inspection template based on the first voice data corresponding to the medical consultant, and quality inspecting the second voice data according to this voice quality inspection template to give real-time reminders to the medical customer service personnel, so as to intervene in the medical service voice process with better comfort, and effectively assist the medical customer service personnel to provide high-quality and standardized medical voice services to the medical consultant. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work.

[0047] Figure 1 It is a schematic diagram of the steps of a medical voice service quality inspection method provided by an embodiment of the present invention;

[0048] Figure 2 It is a schematic diagram of an operation scenario of a medical voice service quality inspection method provided by an embodiment of the present invention;

[0049] Figure 3 It is a schematic diagram of a scenario of identifying first voice data and second voice data in a medical voice service quality inspection method provided by an embodiment of the present invention;

[0050] Figure 4 yes Figure 1 A flowchart of the step of identifying the first voice data and the second voice data in the medical voice service quality inspection method;

[0051] Figure 5 yes Figure 1 A flowchart of the synchronous quality inspection operation steps in the medical voice service quality inspection method;

[0052] Figure 6 A schematic diagram of the module structure of a medical voice service quality inspection device provided by an embodiment of the present invention;

[0053] Figure 7 A schematic block diagram of the structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0054] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0055] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0056] It should be understood that the terms used in this specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0057] It should be further understood that the term “and / or” used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0058] Smart healthcare is a recently emerging medical term that uses the most advanced Internet of Things technology to achieve interaction between patients and medical staff, medical institutions, and medical equipment, gradually achieving informatization. Among them, an important application of smart healthcare is that users can consult medical staff through telephone or other voice communication, thereby realizing online consultation.

[0059] However, in the process of voice communication between medical staff and users, there is a lack of a voice quality inspection system that is closely integrated with medical services and supporting effective intervention measures, which is not conducive to improving the standardization of services provided by relevant medical staff and helping patients build trust in online consultations.

[0060] Based on this, the embodiments of the present invention provide a medical voice service quality inspection method, device, computer equipment and computer-readable storage medium, which are intended to perform quality inspection on medical voice services to provide real-time reminders to medical customer service personnel, and help provide high-quality and standardized medical voice services.

[0061] Among them, the medical voice service quality inspection method can be applied to electronic devices such as computers, intelligent robots, independent control terminals or manager clusters, and is not limited here.

[0062] In this embodiment, the medical voice service quality inspection method is applied to a control terminal as an example for explanation, but the medical voice service quality inspection method is not limited to being used only for a control terminal.

[0063] Some implementation modes of the present invention are described in detail below in conjunction with the accompanying drawings. The following embodiments and features in the embodiments may be combined with each other unless there is any conflict.

[0064] See also Figure 1 and Figure 2 , Figure 1 is a schematic diagram of a process flow of a medical voice service quality inspection method provided by an embodiment of the present invention. Figure 2 It is a schematic diagram of an operation scenario of a medical voice service quality inspection method provided by an embodiment of the present invention. The method specifically includes the following steps S1 to S6.

[0065] Step S1: When a medical voice service quality inspection instruction is received, the medical service voice process corresponding to the medical voice service quality inspection instruction is determined, and the medical service voice process is monitored to obtain real-time voice data.

[0066] Specifically, the medical service voice process refers to: the medical consultant requests medical voice service from the medical customer service staff through the telephone or other voice communication tools, and the medical customer service staff gives corresponding responses based on the consultation of the medical consultant. Among them, the medical customer service staff can be: medical staff, medical institution customer service staff or other medical equipment technicians.

[0067] Exemplarily, the control device for executing the present method may be an operation terminal corresponding to the medical consultant or an operation terminal corresponding to the medical customer service staff to form a communication connection, thereby providing hardware support for monitoring the medical service voice process.

[0068] Specifically, the medical voice service quality inspection instruction may be: when the medical consultant dials the phone of the medical customer service personnel or establishes a connection through other voice communication tools, at least one of the operation terminals corresponding to the medical consultant or the operation terminals corresponding to the medical customer service personnel generates and sends to the control device that executes the present method, that is, starts monitoring when the phone is dialed, thereby promptly starting to pay attention to the medical service voice process and intervene to remind, and obtain complete real-time voice data.

[0069] Step S2: Recognize the real-time voice data to obtain first voice data corresponding to the medical consultant and second voice data corresponding to the medical customer service staff.

[0070] It should be understood that before performing quality inspection on real-time voice data, it is necessary to clearly distinguish the first voice data corresponding to the medical consultant and the second voice data corresponding to the medical customer service staff to avoid confusing the voice content of the two, thereby causing misjudgment or omission during quality inspection.

[0071] See also Figure 3 and Figure 4 , Figure 3 is a schematic diagram of a scenario of identifying first voice data and second voice data in a medical voice service quality inspection method provided by an embodiment of the present invention, Figure 4 yes Figure 1 A flowchart of the step of identifying the first voice data and the second voice data in the medical voice service quality inspection method.

[0072] like Figure 3 and Figure 4 As shown, in some embodiments, step S2 recognizes the real-time voice data to obtain the first voice data corresponding to the medical consultant and the second voice data corresponding to the medical customer service staff, including steps S21 to S23:

[0073] Step S21, inputting the real-time speech data into a preset speech conversion model for text conversion to generate a plurality of sentence segment text information, and each sentence segment text information is respectively associated with a different label time;

[0074] Step S22: using a preset voice feature recognition model to identify the voice features corresponding to each label time in the real-time voice data;

[0075] Step S23: determining first voice data and second voice data from the real-time voice data according to the voice text information and the voice features.

[0076] Specifically, the label time is a period of time on the timeline of real-time voice data, and the segment text information is the corresponding segment text in the real-time voice data within the label time, wherein the label time corresponding to multiple segment text information can be separated according to semantics and / or interval time.

[0077] Specifically, the speech feature corresponding to each label time refers to: the feature of the speech data within the label time, which exemplarily includes but is not limited to: at least one of a voiceprint feature and an intonation feature.

[0078] In some implementations, step S22 uses a preset speech feature recognition model to recognize speech features corresponding to each label time in the real-time speech data, including:

[0079] Extract corresponding sentence segment voice data from real-time voice data according to the label time;

[0080] The sentence speech data is input into the speech feature recognition model for speech feature recognition to generate corresponding speech features.

[0081] Specifically, the segment voice data refers to the data of the voice communication between the medical consultant and the medical customer service personnel during the label time, which can be intercepted from the real-time voice data according to the label time, and the voice features of the segment voice data include but are not limited to at least one of the voiceprint feature and the intonation feature. It should be understood that the voice content of the medical consultant and the voice content of the medical customer service personnel have different voiceprint features and may also have different intonation features.

[0082] In some implementations, step S23 determines the first voice data and the second voice data from the real-time voice data according to the voice text information and the voice feature, including:

[0083] Determine the probability of a first sound source corresponding to the label time based on the voice text information;

[0084] Determine the probability of a second sound source corresponding to the label time based on the voice feature;

[0085] The sound source corresponding to the label time is determined to be a medical consultant or a medical customer service staff according to the first sound source probability and the second sound source probability, so as to determine the first voice data and the second voice data from the real-time voice data.

[0086] Specifically, after determining the voice text information corresponding to the label time and the voice features corresponding to the label time, the corresponding sentence voice data can be judged based on the corresponding voice text information and voice features to determine whether the source of the sentence voice data is a medical consultant or a medical customer service personnel. The above judgment steps are performed on different sentence voice data in the real-time voice data in turn, and the first voice data corresponding to the medical consultant and the second voice data corresponding to the medical customer service personnel can be determined from the real-time voice data.

[0087] The first sound source probability at least includes: the probability of the corresponding sentence voice data coming from the medical consultant and the probability of the corresponding sentence voice data coming from the medical customer service staff based on the voice text information. Similarly, the second sound source probability at least includes: the probability of the corresponding sentence voice data coming from the medical consultant and the probability of the corresponding sentence voice data coming from the medical customer service staff based on the voice features.

[0088] Exemplarily, after determining the first sound source probability and the second sound source probability of the sentence speech data, the total sound source probability of the sentence speech data can be calculated by combining the first sound source probability and the second sound source probability, thereby judging whether the sentence speech data comes from a medical consultant or a medical customer service staff according to the total sound source probability.

[0089] Exemplarily, a weighted average of the first sound source probability and the second sound source probability may be calculated as the total sound source probability, and the sum of the weighted coefficients corresponding to the first sound source probability and the second sound source probability is 1.

[0090] For example, if the probability of the first sound source is: "Medical consultant 70%, medical customer service staff 30%", and the probability of the second sound source is: "Medical consultant 50%, medical customer service staff 50%", the weighting coefficient of the first sound source probability is 0.5, and the weighting coefficient of the second sound source probability is 0.5, then the total sound source probability is: "Medical consultant 60%, medical customer service staff 40%"

[0091] It should be understood that the speech conversion model and the speech feature recognition model are used to divide the real-time speech data into sentence-segment speech data, and determine the speech text information and speech features of the sentence-segment speech data. The two are then combined to accurately distinguish the first speech data corresponding to the medical consultant and the second voice data corresponding to the medical customer service personnel, so as to avoid confusing the speech content of the two and causing misjudgment or missed judgment during quality inspection.

[0092] Step S3: retrieve a matching voice quality inspection template according to the first voice data.

[0093] It should be noted that the first voice data can be used to reflect the medical problems consulted by the medical consultant and / or the tone of voice during the consultation. It should be understood that for different medical problems, especially for consultations corresponding to different departments (such as surgery, dermatology, urology, etc.), it is appropriate to use the voice quality inspection template corresponding to the medical problem to conduct quality inspection and assistance on the speech content of the medical customer service personnel. On the other hand, the consulting tone of the consultant reflects the mood of the medical consultant. In particular, when the medical consultant consults in a more anxious tone, the medical customer service personnel need to pay special attention to ensure calmness and politeness. Therefore, for different consulting tones, the medical customer service personnel are suitable to use the voice quality inspection template corresponding to the consulting tone to conduct quality inspection and assistance on the speech content of the medical customer service personnel.

[0094] Specifically, the first voice data can be analyzed to determine the corresponding medical problem and / or consulting tone, so as to retrieve the voice quality inspection template corresponding to the medical problem and / or consulting tone, thereby realizing humanized and personalized quality inspection, and better helping medical customer service personnel to respond to the medical consultant's questions appropriately and in a standardized manner.

[0095] Step S4: parsing the speech quality inspection template to determine the target intonation and target key words and phrases corresponding to the speech quality inspection template.

[0096] Specifically, the speech quality inspection template includes target intonation and target key words and phrases. The target intonation refers to the intonation determined based on the current medical consultant's consultation content (such as medical questions and consultation tone) to guide the medical consultant to respond with the target intonation, and the target key sentences refer to the terms and sentences determined based on the current medical consultant to guide the medical consultant to mention the above-mentioned target key sentences when responding.

[0097] For example, the target key sentence may be a treatment plan, consultation advice or other sentence corresponding to the medical problem, so as to prevent the medical customer service staff from being unable to remember the correct response for a while.

[0098] Step S5: Perform a synchronous quality inspection operation on the second voice data according to the target intonation and the target key words and phrases to generate real-time reminder information, and send the real-time reminder information to the operation terminal of the medical customer service personnel.

[0099] See also Figure 5 , Figure 5 yes Figure 1 A flowchart of the synchronous quality inspection operation steps in the medical voice service quality inspection method.

[0100] like Figure 5 As shown, in some embodiments, the real-time reminder information includes at least one of intonation reminder information and key word reminder information. In step S5, a synchronous quality inspection operation is performed on the second voice data according to the target intonation and the target key word, including:

[0101] Step S51: identifying a first real-time intonation of the first voice data and a second real-time intonation of the second voice data; and generating intonation reminder information according to the target intonation, the first real-time intonation and the second real-time intonation.

[0102] It should be understood that after the target tone is determined, the target tone can be adjusted in real time according to the tone changes of the medical consultant over time.

[0103] Specifically, generating the intonation reminder information according to the target intonation, the first real-time intonation and the second real-time intonation includes: dynamically adjusting the target intonation according to the first real-time intonation to obtain the real-time target intonation, calculating the matching degree between the real-time target intonation and the second real-time intonation, and then generating the intonation reminder information according to the matching degree.

[0104] Exemplarily, generating tone reminder information based on the degree of matching may be: generating tone reminder information when the degree of matching is less than a preset matching threshold to prompt the medical customer service personnel; not generating tone reminder information when the degree of matching is greater than or equal to the preset matching threshold, and not interfering with the normal response of the medical customer service personnel.

[0105] Exemplarily, the speech feature recognition model may be used to recognize the intonation feature of the first speech data as the first real-time intonation, and / or recognize the intonation feature of the second speech data as the second real-time intonation.

[0106] Step S52: determining the frequency information of the target key words and phrases in the target key words and phrases, and generating key word and phrase reminder information according to the frequency information.

[0107] In some implementations, the target keyword phrase includes at least one target sub-word phrase; generating keyword phrase reminder information according to the frequency information in step S52 includes:

[0108] Determine the occurrence frequency corresponding to the target sub-word sentence according to the frequency information;

[0109] The occurrence frequency is compared with the target frequency to generate sub-reminder information corresponding to the target sub-word and sentence, and the sub-reminder information is integrated to generate the word and sentence reminder information, wherein the target frequency is associated with the target sub-word and sentence.

[0110] Specifically, the occurrence frequency is compared with the target frequency to generate sub-reminder information corresponding to the target sub-word, including: when the occurrence frequency is less than the target frequency, the sub-reminder information corresponding to the target sub-word is generated to prompt the medical customer service personnel; when the occurrence frequency is greater than or equal to the target frequency, the sub-reminder information is not generated, and the normal response of the medical customer service personnel is not interfered with.

[0111] like Figure 2As shown, specifically, sending tone reminder information and / or sending key word reminder information can be: sending the tone reminder information and / or key word reminder information to the operating terminal of the medical customer service personnel to generate a corresponding prompt signal at the operating terminal of the medical customer service personnel, thereby instantly reminding the medical customer service personnel to adjust the response tone or use corresponding sentences.

[0112] Exemplarily, the operating terminal of the medical customer service staff is, for example, a computer, a smart phone, a smart watch, etc.

[0113] Step S6: When the preset scoring conditions are met, a quality inspection score of the medical service process is generated according to the first voice data and the second voice data.

[0114] Exemplarily, the preset scoring conditions may be: the medical service voice process is completed, and / or the medical service voice process is performed every preset time.

[0115] It should be understood that the quality inspection score of the medical service process generated based on the first voice data and the second voice data is a summary quality inspection score of the medical service process after a period of time or after the end of the process, rather than on-the-spot intervention in the medical service process between the medical consultant and the medical customer service personnel, which is different from the synchronous quality inspection operation performed in real-time to assist the medical service voice process.

[0116] In some embodiments, generating a quality inspection score of the medical service process according to the first voice data and the second voice data in step S6 includes:

[0117] Performing semantic recognition on the first voice data to obtain consultation content information corresponding to the first voice data;

[0118] Extracting keywords from the second voice data to obtain answer keywords;

[0119] The accuracy of the medical customer service staff's responses is determined based on the consultation content information and response keywords, thereby generating a quality inspection score.

[0120] Specifically, semantic recognition is performed on the first voice data to determine the actual consultation content information of the medical consultant, including medical questions and / or consultation tone, and then a standard response template is generated based on the consultation content information. The response keywords in the second voice data and the standard response template are further used to determine whether the medical customer service staff's response is accurate and standardized, and whether it provides the medical customer service staff with a correct diagnosis or psychological guidance, thereby giving a quality inspection score.

[0121] In summary, the embodiments of the present invention provide a medical voice service quality inspection method, apparatus, computer equipment and computer-readable storage medium, wherein the medical voice service quality inspection method includes: when a medical voice service quality inspection instruction is received, determining the medical service voice process corresponding to the medical voice service quality inspection instruction, and monitoring the medical service voice process to obtain real-time voice data; identifying the real-time voice data to obtain first voice data corresponding to the medical consultant and second voice data corresponding to the medical customer service personnel; retrieving a matching voice quality inspection template according to the first voice data; parsing the voice quality inspection template to determine the target intonation and target key words and phrases corresponding to the voice quality inspection template; and retrieving the matching voice quality inspection template according to the target intonation and target key words and phrases. The key words and phrases perform synchronous quality inspection operations on the second voice data to generate real-time reminder information, and send the real-time reminder information to the operation terminal of the medical customer service personnel; when the preset scoring conditions are met, a quality inspection score of the medical service process is generated according to the first voice data and the second voice data, by monitoring the medical service voice process, and retrieving the matching voice quality inspection template based on the first voice data corresponding to the medical consultant, and performing quality inspection on the second voice data according to this voice quality inspection template to give real-time reminders to the medical customer service personnel, so as to intervene in the medical service voice process with better comfort, and effectively assist the medical customer service personnel to provide high-quality and standardized medical voice services to the medical consultant.

[0122] The embodiment of the present invention also provides a medical voice service quality inspection device, Figure 6 A schematic diagram of the module structure of a medical voice service quality inspection device provided by an embodiment of the present invention, such as Figure 6 As shown, the medical voice service quality inspection device 700 includes:

[0123] The voice monitoring module 701 is used to determine the medical service voice process corresponding to the medical voice service quality inspection instruction when receiving the medical voice service quality inspection instruction, and monitor the medical service voice process to obtain real-time voice data;

[0124] A sound source recognition module 702 is used to recognize the real-time voice data to obtain first voice data corresponding to the medical consultant and second voice data corresponding to the medical customer service staff;

[0125] The template retrieval module 703 is used to retrieve a matching voice quality inspection template according to the first voice data;

[0126] The template parsing module 704 is used to parse the speech quality inspection template to determine the target intonation and target key words and phrases corresponding to the speech quality inspection template;

[0127] The first quality inspection module 705 is used to perform a synchronous quality inspection operation on the second voice data according to the target intonation and the target key words and phrases to generate real-time reminder information, and send the real-time reminder information to the operation terminal of the medical customer service personnel;

[0128] The second quality inspection module 706 is used to generate a quality inspection score for the medical service process according to the first voice data and the second voice data when a preset scoring condition is met.

[0129] In some implementations, the sound source recognition module 702 recognizes the real-time voice data to obtain first voice data corresponding to the medical consultant and second voice data corresponding to the medical customer service staff, including:

[0130] Inputting the real-time speech data into a preset speech conversion model for text conversion to generate a plurality of segment text information, wherein each segment text information is respectively associated with a different label time;

[0131] The preset speech feature recognition model is used to identify the speech features corresponding to each label time in the real-time speech data;

[0132] The first voice data and the second voice data are determined from the real-time voice data according to the voice text information and the voice feature.

[0133] In some implementations, the sound source identification module 702 uses a preset speech feature recognition model to identify speech features corresponding to each label time in the real-time speech data, including:

[0134] Extract corresponding sentence segment voice data from real-time voice data according to the label time;

[0135] The sentence speech data is input into the speech feature recognition model for speech feature recognition to generate corresponding speech features.

[0136] In some implementations, the sound source identification module 702 determines the first voice data and the second voice data from the real-time voice data according to the voice text information and the voice features, including:

[0137] Determine the probability of a first sound source corresponding to the label time based on the voice text information;

[0138] Determine the probability of a second sound source corresponding to the label time based on the voice feature;

[0139] The sound source corresponding to the label time is determined to be a medical consultant or a medical customer service staff according to the first sound source probability and the second sound source probability, so as to determine the first voice data and the second voice data from the real-time voice data.

[0140] In some implementations, the real-time reminder information includes at least one of intonation reminder information and key word reminder information. The first quality inspection module 705 performs a synchronous quality inspection operation on the second voice data according to the target intonation and the target key word, including:

[0141] Recognize a first real-time intonation of the first voice data and a second real-time intonation of the second voice data; generate intonation reminder information according to the target intonation, the first real-time intonation and the second real-time intonation;

[0142] and / or,

[0143] Determine the frequency information of the target key words and phrases in the target key words and phrases, and generate key word and phrase reminder information based on the frequency information.

[0144] In some implementations, the target keyword phrase includes at least one target sub-word phrase; the first quality inspection module 705 generates keyword phrase reminder information according to the frequency information, including:

[0145] Determine the occurrence frequency corresponding to the target sub-word sentence according to the frequency information;

[0146] The occurrence frequency is compared with the target frequency to generate sub-reminder information corresponding to the target sub-word and sentence, and the sub-reminder information is integrated to generate the word and sentence reminder information, wherein the target frequency is associated with the target sub-word and sentence.

[0147] In some implementations, the first quality inspection module 706 generates a quality inspection score for the medical service process based on the first voice data and the second voice data, including:

[0148] Performing semantic recognition on the first voice data to obtain consultation content information corresponding to the first voice data;

[0149] Extracting keywords from the second voice data to obtain answer keywords;

[0150] The accuracy of the medical customer service staff's responses is determined based on the consultation content information and response keywords, thereby generating a quality inspection score.

[0151] The embodiment of the present invention also provides a computer device, see Figure 7 , Figure 7 A schematic block diagram of the structure of a computer device provided in an embodiment of the present invention.

[0152] like Figure 7 As shown, the computer device 800 includes a processor 801 and a memory 802 , and the processor 801 and the memory 802 are connected via a bus 803 , such as an I2C (Inter-integrated Circuit) bus.

[0153] Specifically, the processor 801 is used to provide computing and control capabilities to support the operation of the entire computer device. The processor 801 can be a central processing unit (CPU), and the processor 801 can also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0154] Specifically, the memory 802 may be a Flash chip, a read-only memory (ROM) disk, an optical disk, a USB flash drive, or a mobile hard disk.

[0155] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the embodiment of the present invention, and does not constitute a limitation on the computer device to which the embodiment of the present invention is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0156] The processor is used to run a computer program stored in the memory, and implement any one of the medical voice service quality inspection methods provided by the embodiments of the present invention when executing the computer program.

[0157] In some implementations, the processor 801 is configured to run a computer program stored in the memory 802, and implement the following steps when executing the computer program:

[0158] When receiving a medical voice service quality inspection instruction, determining the medical service voice process corresponding to the medical voice service quality inspection instruction, and monitoring the medical service voice process to obtain real-time voice data;

[0159] Recognize the real-time voice data to obtain first voice data corresponding to the medical consultant and second voice data corresponding to the medical customer service staff;

[0160] Retrieving a matching voice quality inspection template according to the first voice data;

[0161] Parse the speech quality inspection template to determine the target intonation and target key words and phrases corresponding to the speech quality inspection template;

[0162] Performing a synchronous quality inspection operation on the second voice data according to the target intonation and the target key words and phrases to generate real-time reminder information, and sending the real-time reminder information to the operation terminal of the medical customer service personnel;

[0163] When the preset scoring conditions are met, a quality inspection score of the medical service process is generated according to the first voice data and the second voice data.

[0164] In some implementations, the processor 801 recognizes the real-time voice data to obtain first voice data corresponding to the medical consultant and second voice data corresponding to the medical customer service staff, including:

[0165] Inputting the real-time speech data into a preset speech conversion model for text conversion to generate a plurality of segment text information, wherein each segment text information is respectively associated with a different label time;

[0166] The preset speech feature recognition model is used to identify the speech features corresponding to each label time in the real-time speech data;

[0167] The first voice data and the second voice data are determined from the real-time voice data according to the voice text information and the voice feature.

[0168] In some implementations, the processor 801 uses a preset voice feature recognition model to recognize voice features corresponding to each label time in the real-time voice data, including:

[0169] Extract corresponding sentence segment voice data from real-time voice data according to the label time;

[0170] The sentence speech data is input into the speech feature recognition model for speech feature recognition to generate corresponding speech features.

[0171] In some implementations, the processor 801 determines the first voice data and the second voice data from the real-time voice data according to the voice text information and the voice feature, including:

[0172] Determine the probability of a first sound source corresponding to the label time based on the voice text information;

[0173] Determine the probability of a second sound source corresponding to the label time based on the voice feature;

[0174] The sound source corresponding to the label time is determined to be a medical consultant or a medical customer service staff according to the first sound source probability and the second sound source probability, so as to determine the first voice data and the second voice data from the real-time voice data.

[0175] In some implementations, the processor 801 real-time reminder information includes at least one of intonation reminder information and key word reminder information, and performs a synchronous quality inspection operation on the second voice data according to the target intonation and the target key word, including:

[0176] Recognize a first real-time intonation of the first voice data and a second real-time intonation of the second voice data; generate intonation reminder information according to the target intonation, the first real-time intonation and the second real-time intonation;

[0177] and / or,

[0178] Determine the frequency information of the target key words and phrases in the target key words and phrases, and generate key word and phrase reminder information based on the frequency information.

[0179] In some implementations, the processor 801 generates keyword reminder information based on the frequency information, including:

[0180] Determine the occurrence frequency corresponding to the target sub-word sentence according to the frequency information;

[0181] The occurrence frequency is compared with the target frequency to generate sub-reminder information corresponding to the target sub-word and sentence, and the sub-reminder information is integrated to generate the word and sentence reminder information, wherein the target frequency is associated with the target sub-word and sentence.

[0182] In some implementations, the processor 801 generates a quality inspection score of the medical service process according to the first voice data and the second voice data, including:

[0183] Performing semantic recognition on the first voice data to obtain consultation content information corresponding to the first voice data;

[0184] Extracting keywords from the second voice data to obtain answer keywords;

[0185] The accuracy of the medical customer service staff's responses is determined based on the consultation content information and response keywords, thereby generating a quality inspection score.

[0186] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the computer device described above can refer to the corresponding process in the aforementioned animal identification method embodiment, and will not be repeated here.

[0187] An embodiment of the present invention also provides a computer-readable storage medium, which stores a computer program. The computer program can be executed by one or more processors to implement the steps of any medical voice service quality inspection method provided in the embodiments of this specification.

[0188] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc., equipped on the computer device.

[0189] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In a hardware embodiment, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0190] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0191] The serial numbers of the embodiments of the present invention are only for description and do not represent the advantages and disadvantages of the embodiments. The above description is only a specific embodiment of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention shall be based on the protection scope of the claims.

Claims

1. A medical voice service quality inspection method, characterized in that: The method comprises: When a medical voice service quality inspection instruction is received, determining a medical service voice process corresponding to the medical voice service quality inspection instruction, and monitoring the medical service voice process to obtain real-time voice data; Recognize the real-time voice data to obtain first voice data corresponding to the medical consultant and second voice data corresponding to the medical customer service staff; Retrieving a matching voice quality inspection template according to the first voice data; Parsing the speech quality inspection template to determine the target intonation and target key words and phrases corresponding to the speech quality inspection template; Performing a synchronous quality inspection operation on the second voice data according to the target intonation and the target key words and phrases to generate real-time reminder information, and sending the real-time reminder information to the operation terminal of the medical customer service staff; When a preset scoring condition is met, a quality inspection score of the medical service process is generated according to the first voice data and the second voice data.

2. The method according to claim 1, characterized in that The identifying the real-time voice data to obtain first voice data corresponding to the medical consultant and second voice data corresponding to the medical customer service staff includes: Inputting the real-time voice data into a preset voice conversion model for text conversion to generate a plurality of sentence segment text information, wherein each sentence segment text information is respectively associated with a different label time; Using a preset speech feature recognition model to identify the speech features corresponding to each of the label times in the real-time speech data; The first voice data and the second voice data are determined from the real-time voice data according to the voice text information and the voice feature.

3. The method according to claim 2, characterized in that The using a preset speech feature recognition model to identify the speech features corresponding to each of the label times in the real-time speech data includes: Extracting corresponding sentence segment voice data from the real-time voice data according to the label time; The sentence segment speech data is input into the speech feature recognition model for speech feature recognition to generate the corresponding speech features.

4. The method according to claim 2, characterized in that The determining the first voice data and the second voice data from the real-time voice data according to the voice text information and the voice feature includes: Determining a first sound source probability corresponding to the label time based on the voice text information; Determining a second sound source probability corresponding to the label time based on the voice feature; The sound source corresponding to the label time is determined to be the medical consultant or the medical customer service staff according to the first sound source probability and the second sound source probability, so as to determine the first voice data and the second voice data from the real-time voice data.

5. The method according to claim 1, characterized in that The real-time reminder information includes at least one of intonation reminder information and key word reminder information, and the synchronous quality inspection operation on the second voice data according to the target intonation and the target key word includes: Identify a first real-time intonation of the first voice data and a second real-time intonation of the second voice data; generate the intonation reminder information according to the target intonation, the first real-time intonation and the second real-time intonation; and / or, The frequency information of the target key words and phrases in the target key words and phrases is determined, and the key word and phrase reminder information is generated according to the frequency information.

6. The method according to claim 5, characterized in that The target keyword sentence includes at least one target sub-word sentence; and generating the keyword sentence reminder information according to the frequency information includes: Determining the occurrence frequency corresponding to the target sub-word sentence according to the frequency information; The occurrence frequency is compared with a target frequency to generate sub-reminder information corresponding to the target sub-word and sentence, and the word and sentence reminder information is generated by integrating the sub-reminder information, wherein the target frequency is associated with the target sub-word and sentence.

7. The method according to claim 1, characterized in that Generating a quality inspection score for the medical service process according to the first voice data and the second voice data includes: Performing semantic recognition on the first voice data to obtain consultation content information corresponding to the first voice data; Extracting keywords from the second voice data to obtain answer keywords; The accuracy of the medical customer service personnel's response is determined based on the consultation content information and the response keywords, thereby generating the quality inspection score.

8. A medical voice service quality inspection device, characterized in that: include: A voice monitoring module is used to determine the medical service voice process corresponding to the medical voice service quality inspection instruction when receiving the medical voice service quality inspection instruction, and monitor the medical service voice process to obtain real-time voice data; A sound source recognition module, used to recognize the real-time voice data to obtain first voice data corresponding to the medical consultant and second voice data corresponding to the medical customer service staff; A template retrieval module, used to retrieve a matching voice quality inspection template according to the first voice data; A template parsing module, used to parse the speech quality inspection template to determine the target intonation and target key words and phrases corresponding to the speech quality inspection template; A first quality inspection module, configured to perform a synchronous quality inspection operation on the second voice data according to the target intonation and the target key words and phrases to generate real-time reminder information, and send the real-time reminder information to the operation terminal of the medical customer service personnel; The second quality inspection module is used to generate a quality inspection score for the medical service process according to the first voice data and the second voice data when a preset scoring condition is met.

9. A computer device, characterized in that: The computer device includes a memory and a processor; The memory is used to store computer programs; The processor is used to execute the computer program and implement the medical voice service quality inspection query method as described in any one of claims 1 to 7 when executing the computer program.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor implements the medical voice service quality inspection method according to any one of claims 1 to 7.

Images