33 results about "Voice of the customer" patented technology

Chat categorization uses semi-supervised clustering to provide Voice of the Customer (VOC) analytics over unstructured data via an historical understanding of topic categories discussed to derive an automated methodology of topic categorization for new data; application of semi-supervised clustering (SSC) for VOC analytics; generation of seed data for SSC; and a voting algorithm for use in the absence of domain knowledge / manual tagged data. Customer service interactions are mined and quality of these interactions is measured by “Customer's Vote” which, in turn, is determined by the customer's experience during the interaction and the quality of customer issue resolution. Key features of the interaction that drive a positive experience and resolution are automatically learned via machine learning driven algorithms based on historical data. This, in turn, is used to coach / teach the system / service representative on future interactions.

The invention discloses a conversation voice monitoring method and system. The method comprises: collecting voice data of conversion of a server side and a client in real time; carrying out voice identification on the voice data of the service side and the client to obtain a server side identification text and a client identification text; and according to the conversation voice data of the server side and the corresponding identification text as well as the conversation voice data of the client and the corresponding identification text, carrying out real-time monitoring on the conversation voice of the server side. With the method and system, comprehensive and real-time monitoring of conversation voices of the customer service staff can be realized; and the service quality of the customer service staff is improved.

The invention discloses a method for automatically detecting customer service voices, which comprises the following steps of: generating a test script according to a preset test task; detecting a detected customer service center according to the test script; and processing a detection result. The invention also discloses a system for automatically detecting the customer service voices. By automatically detecting the voices of the customer service center, the method and the system of the invention obtain the current working condition of the customer service center for servicers to manage, thereby realizing automatization, batch and customization of the detecting process of the customer service center and reducing the error probability of the detection and the labor cost.

The invention relates to the technical field of information processing, and discloses a customer service agent emotion monitoring system. The system comprises a face emotion monitoring module, a voiceemotion monitoring module, a logic comparison module and an input module, which are connected with a processor, wherein the logic comparison module is furthermore connected with an alarm module; theinput module is used for inputting voice emotion values and face emotion values of one or more to-be-monitored emotion types; the face emotion monitoring module is used for analyzing face videos of customer service agents and outputting different types of face emotion values of the customer service agents; the voice emotion monitoring module is used for analyzing voices of the customer service agents and outputting different types of voice emotion values of the customer service agents; the logic comparison module is used for recognizing a corresponding relationship between the customer serviceagents and multiple different types of face emotion values and voice emotion values, comparing the multiple different face emotion values and voice emotion values with the voice emotion values and face emotion values of the to-be-monitored emotion types, and outputting alarm information; and the alarm module is used for receiving the alarm information and giving alarms.

The present invention proposes a product scoring method and device based on customer voice VOC, the product scoring method based on customer voice VOC includes determining a plurality of performance parameters of the product to be rated, and determining a plurality of scoring items for scoring the product to be rated; Multiple performance parameters and multiple scoring items generate a fuzzy transformation model, and determine the scoring model corresponding to the fuzzy transformation model; determine the target value of each performance parameter from multiple technical target values ​​of the product to be scored based on the ant colony algorithm; The target value of each performance parameter and the scoring model are used to score the product to be rated. The invention can effectively improve the objectivity and referenceability of product scoring in the technical field of automobile engineering, and improve the scoring effect.

The invention discloses a three-dimension voice recognition method for smart canteen card readers, aiming to solve the technical problem of low settlement efficiency of existing canteen card readers. The technical scheme includes utilizing voice sensors A1, A2, A3 and O to receive voice of the same customer at different time, then calculating position coordinates of the customer according to difference of times of receiving the voice; judging whether the position of the customer is within a predefined effective range or not; submitting the voice information to a card reader server through a card reader main controller, calculating the total amount of consumption of the customer, feeding the total amount back to a card reader terminal and displaying the total amount on an LED display screen. Whether the customer is within the effective range or not is judged according to a three-dimension voice source; if the voice source is effective, then the voice of the customer is submitted to the card reader server, and effective required information is extracted through voice recognition software of the card reader server, and the required information and the corresponding total amount of consumption are fed back to the card reader terminal. By the three-dimension voice recognition method, positions of customers can be recognized accurately and settlement efficiency of the canteen card readers is improved.

An intelligent customer return visit method includes obtain transaction record, extracting customer information according to that transaction record; according to the extracted customer information, automatically calling the network telephone to send a call request for establishing a telephone connection to the customer; when the telephone connection is established successfully, acquiring and parsing the voice content of the customer; according to the keywords included in the parsed voice content, the voice matching the voice meaning of the keywords is searched and transmitted in the pre-recorded voice, and the feedback information of the customer is recorded, so that the customer return visit can be completed automatically by the system. There is no need to train or hire special customerservice personnel, which greatly reduces the labor cost and time cost, and the return visit process is recorded according to the voice of the customer adaptive response and objectively and is not be affected by the personal subjective factors of the return visit personnel, which is conducive to improve the quality of the return visit.

The invention discloses a sound optimization system. The sound optimization system comprises a console, a processor and a server; the console is connected to the server; the processor is connected tothe console; the console is used for setting and adjusting sound optimization parameters; the processor is used for performing sound optimization processing according to the set and adjusted sound optimization parameters; and the server is used for storing a recording result, the sound optimization parameter and the corresponding sound optimization mode. In addition, the invention also discloses asound optimization method. By the adoption of the voice optimization method and system, a manual service agent can call the voice optimization mode to optimize the voice of a complaint client, the working pressure of the complaint client is effectively relieved, and the working experience is improved; when a voice robot transfers the manual service agent, the tone and the timbre of the manual service agent are consistent and coherent with the voice robot, a specific voice optimization mode can be selected and configured for the voice robot according to customer attributes, the tone and the timbre of the manual service agent are converted into a target optimization mode, and the customer experience is improved.

A method and system for voice emotion identification contained in audio in a call providing customer support between a customer and a service agent by implementing an emotion identification application to identify emotions captured in a voice of the customer from audio received by a media streaming device; receiving, by the emotion identification application, an audio stream of a series of voice samples contained in consecutive frames from audio received; extracting, by the emotion identification application, a set of voice emotion features from each frame in each voice sample of the audio by applying a trained machine learning (ML) model for identifying emotions utilizing a neural networks to determine one or more voice emotions by a configured set of voice emotion features captured in each voice sample; and classifying, by the emotion identification application, each emotion determined by the trained ML model based on a set of classifying features to label one or more types of emotions captured in each voice sample.

The invention relates to a consultation recommendation method and a management system thereof. The method comprises the steps that voice of a customer is acquired through a voice acquisition module when a commodity service is recommended to the customer; the volume, speed and tone of the voice are analyzed in real time according to the voice of the customer, and the real-time emotional state of the customer is determined according to volume information, speed information and tone information; the interest degree of the customer is analyzed according to the real-time emotional state of the customer; customer service staff make further recommendation according to the recommendation content which the customer is interested in. The system comprises an acquisition module, a voice analysis module and a emotion analysis module, wherein the acquisition module is used for acquiring the voice emitted by the customer when consulting the customer service staff; the voice analysis module is used for analyzing the volume, speed and tone of the voice of the customer, determining the changing range of the voice of the customer, and detecting whether the voice of the customer exceeds the designatedarea or not; and the emotion analysis module is used for determining the emotional change of the customer and understanding the recommendation content which the customer is interested in. The customer service staff can understand the emotional state of the customer in real time in providing the consulting service and indirectly improving the consumption desire of the customers.

The invention discloses an insurance man-machine coupling customer service system and method based on artificial intelligence. The system comprises an artificial customer service selection unit, a voice recognition unit, a machine artificial intelligence coupling unit and a voice generation unit. The manual customer service selection unit is used for directly realizing chatting between the customer and the manual customer service; The voice recognition unit converts the voice information of the client into text information and outputs the text information; The machine artificial intelligence coupling unit receives the text information and the information of the manual customer service selection unit at the same time, chatting between a client and a manual customer service or intelligent generation of a natural answer text is achieved, and reinforcement learning of a machine is carried out. And the voice generation unit sends the input natural answer text to the voice synthesis engine,generates voice information and feeds the voice information back to the client. According to the invention, capabilities of pre-sales, after-sales consultation, product recommendation, common knowledge problems and the like can be provided for clients. And the advantages of high accuracy, quick response, high reliability, natural interaction and the like can be achieved at the same time.

The invention provides an intention recognition method and device, electronic equipment and a computer readable storage medium, and the method comprises the steps of obtaining a current voice sent by a customer in an outbound scene, and obtaining a corresponding target text based on the current voice; and determining an intention type included in the target text, and obtaining an intention recognition result corresponding to the target text based on the intention type. According to the scheme, the voice of the customer in the outbound scene is converted into the corresponding target text, then the intention type in the target text is determined, different intention recognition models are adopted for different intention types, the intention of the customer is recognized, and due to the fact that different intention recognition models are adopted for different intention types, the recognition accuracy of voice containing multiple intentions can be improved.

The invention relates to the field of exhibition hall display, and particularly relates to a multimedia intelligent exhibition hall comprising a central processor, a camera, a sound receiver, a 3D projector, a cloud side system and a display screen. The 3D projector projects information, people and slides to be displayed in the 3D form. The crowd is attracted by the 3D projector in actual displayand then the sound receiver collects the sound of the customer and transmits the sound of the customer to the central processor by describing the required data to the sound receiver. The central processor analyzes the received sound and performs matching and comparison with the data in the cloud side system. The central processor calls the successfully matched data to be displayed through the display screen and the 3D projector. Meanwhile, the called data are reread through the sound equipment so that understanding can be facilitated for the crowd, a quite atmospheric way of exhibition can becreated and a great effect of publicity can be achieved.

The invention provides a customer service switching method and device, and the method comprises the steps: connecting an intelligent customer service for a customer when the customer accesses a customer service system, and converting the collected call voice of the customer into a voice signal; performing emotion recognition on the voice signal by applying a preset emotion recognition method, andmarking a corresponding emotion label for the voice signal; determining the emotion level, the early warning level and the early warning frequency of the customer according to the emotion label and the voice portrait of the customer; judging whether the customer meets a preset customer service switching condition or not according to the emotion level, the early warning level and the early warningfrequency; and if yes, switching the customer to the manual customer service, generating early warning information, and updating the early warning information and the emotion label corresponding to the call voice of the customer to the sound portrait. By applying the method and the device, when the emotion of the client is abnormal, the client is switched to the manual customer service in time, sothat the manual customer service can pacify the client, and therefore, high-quality service can be provided for the client.

The invention provides a method and a system for controlling call admission, and the method comprises the following steps: S1, preliminarily determining whether a customer is a valid customer or not by a customer service robot system according to an incoming number, and if the customer is a valid customer, then determining the gender of the customer according to the voice of the customer; S2, simulating the voice of the opposite sex as for the gender of the customer, dialoguing with the customer by utilizing the simulated voice of the opposite sex, collecting the voice of the customer, and analyzing the needs of the customer; S3, searching person information corresponding to the product information according to product information in which the customer is interested, pushing the product information and the associated person avatar to a client terminal for displaying, and dialoguing with the customer by utilizing the voice corresponding to the gender of the person associated with the product at the same time; and S4, determining whether the customer is a valid customer or not again, if the customer is a valid customer, perfecting the customer information, and automatically distributing the customer to a corresponding salesperson according to the needs of the customer.

The invention relates to the field of artificial intelligence, discloses a voice quality inspection method, device and equipment for customer service staff and a storage medium, and is used for improving the scoring accuracy of voice of the customer service staff so as to improve the comprehensiveness and accuracy of voice quality inspection of the customer service staff. The voice quality inspection method for the customer service staff comprises the following steps: acquiring a voice file of the customer service staff, wherein the voice file is a group of complete voice conversations; performing voice recognition on the voice file to obtain a voice vector and a context vector; performing recognition based on the sound vector and the context vector, and generating at least two voiceprint features and text data; based on the context vector, at least two voiceprint features and the text data, carrying out the score calculation on a semantic level and a key information pair level, and obtaining a target comprehensive score, wherein the target comprehensive score is used for representing a voice quality inspection result of the customer service staff.

Embodiments of the invention provide a voice processing method and device, a storage medium and electronic equipment. The voice processing method comprises the steps of detecting a greeting language according to a received first audio signal and responding to the detected greeting language, extracting the voice characteristic of the first audio signal as a reference voice characteristic; extracting the voice characteristic of a second audio signal by responding to the received second audio signal, determining the similarity between the voice characteristics of the second audio signal and the reference voice characteristic; and filtering the second audio signal in response to the similarity being less than a first threshold value, so that the voice of a customer and voice of other people iseffectively distinguished, the problem that when the voice of other people is heard, system also performs voice processing is solved, the error rate of the system is reduced, and the usability of thesystem is enhanced.

The invention relates to an intelligent customer voice insight system. The system comprises a text preprocessing engine, a pre-trained service label recognition model, an emotion recognition model andan entity part recognition model, wherein client voice data is cleaned through a text preprocessing engine, meaningless characters and words are filtered out, then word segmentation and ngram processing are carried out, and word segmentation character variables corresponding to single characters and words and equal-length character variables corresponding to sentences equal in length are obtained, and the word segmentation character variable and the equal-length character variable are respectively inputted into a pre-trained service label identification model, an emotion identification modeland an entity part identification model to obtain a service label, an emotion label, an entity part label and a problem label which correspond to the customer voice data and are identified and confirmed by the models. According to the method, accurate semantic analysis and sentiment analysis can be carried out on the voice of the customer by crawling all comment contents of related channels, so product improvement and service promotion are realized.

The invention relates to the technical field of consultation service, and discloses a consultation service method and a management system thereof.The method comprises the steps that when commodity service is recommended to a client, acquiring voice facial expression information and voice information of the client by an audio and video collection module of a consultation server; preliminarily judging the emotional state of the client according to the facial expression information of the client through facial expression data set detection; analyzing the volume, the speed and the tone of the sound in real time according to the sound information of the client, and further determining the real-time emotional state of the client according to the volume information, the speed information and thetone information; analyzing the degree of interest of the client in each recommended content according to the real-time emotional state of the client; and enabling service staff to accurately recommend according to the recommendation content interested by the client. According to the technical scheme provided by the invention, the service staff can know the emotional state of the client in real time and accurately recommend the commodity service interested by the client, and the order yield is improved.

The invention provides a health code payment method and device based on a 5G message, and belongs to big data. The method comprises the steps of binding a bank card in the 5G message, and determining bank payment information; performing biological information identification on a customer, and determining a biological identification result; uploading the bank payment information and the biological identification result to a 5G message server, comparing the bank payment information and the biological identification result with pre-stored customer personal information, and returning authentication success information to the 5G message when the bank payment information and the biological identification result are consistent with the pre-stored customer personal information; after the authentication success message is received by the 5G message, starting a health code generation function, monitoring the voice of the customer, and determining a voice arousing instruction, wherein the health code generation function is used for generating a health code with a payment function according to bank payment information; and calling a health code generation function according to the voice arousing instruction, and generating a health code with a payment function for payment. According to the invention, the health code is generated in the 5G message and payment is carried out, one code has two purposes, the process is simple, and the customer experience is improved.

The invention is suitable for the technical field of computers, and particularly relates to a voice intelligent customer service text quality inspection method and system, and the method comprises the steps: obtaining a voice interaction record and customer service information; calling corresponding customer service voiceprint information from a preset voiceprint database according to the customer service information; extracting corresponding customer service voice information from the voice interaction record according to the customer service voiceprint information; performing voice recognition on the customer service voice information to obtain a voice recognition result; and evaluating the customer service according to the voice recognition result to generate an evaluation result. According to the voice intelligent customer service text quality inspection method provided by the embodiment of the invention, the voice of the customer service in the voice interaction record is independently extracted according to the voiceprint, semantic analysis and mood analysis are performed on the voice, and the speaking mood of the customer service is judged by using the loudness and accent distribution, so that the voice service provided by the customer service is comprehensively analyzed, and the quality inspection efficiency of the customer service is improved. The behavior of avoiding quality inspection by changing the tone of the customer service staff is avoided, and the accuracy of service quality evaluation is improved.

The invention discloses a method for improving the voice infectivity of a customer service staff. The method for improving the voice infectivity of the customer service staff comprises the following steps of s01, scoring voice of the customer service staff according to feedback evaluation of a user, and selecting first 10% of the voice in scoring as standard voice; s02, extracting an eigenvalue ofthe standard voice and calculating an average value to obtain a standard eigenvalue range; s03, acquiring target voice and carrying out pre-emphasis and framing processing to acquire an eigenvalue which is the same as that of the standard voice; s04, judging whether the eigenvalue of the target voice is in the standard eigenvalue range or not, if so, directly taking the target voice as restored voice, and if not, performing weighting processing to enable the eigenvalue of the target voice to approach the standard eigenvalue range; and s05, restoring voice data according to a weighting processing result in the step s04 to obtain the voice, and outputting the voice. The method has the advantages that the infectivity of the target voice is improved and beautified, and the purpose of improving the voice infectivity of the customer service staff is achieved.

The invention discloses an information mining method, system and device, a storage medium. The information mining method comprises the steps: recognizing a customer based on a face recognition technology, and distributing first identity information to the customer; acquiring activity information of the client in a preset area; if the customer generates the consumption information, allocating a plurality of pieces of second identity information to the customer, and associating the first identity information and / or the second identity information of the customer with a to-be-consulted party of the customer; setting a search instruction according to a preset search classification condition; obtaining dialogue voice of the customer based on the activity information; classifying the second identity information and / or dialogue voice according to a preset classification condition, and mapping the classified second identity information and / or dialogue voice with a search instruction; and if the search instruction is identified, visually displaying the second identity information and / or dialogue voice mapped with the search instruction.

The invention provides a customer post-loan management method, system and device based on a voiceprint recognition technology and a medium, and the method comprises the steps: collecting a sound sample of a customer needing loan, and forming a voiceprint feature of the customer; and according to a set rule, carrying out telephone verification on the customer through the intelligent voice center, judging that a telephone answering person is the customer according to voiceprint characteristics, if so, carrying out corresponding operation, and if not, generating early warning data. A robot information checking, loan marketing and loan repayment reminding mode is formed, the working efficiency is greatly improved, and the working cost is reduced.

The invention discloses a conversation voice monitoring method and system. The method comprises: collecting voice data of conversion of a server side and a client in real time; carrying out voice identification on the voice data of the service side and the client to obtain a server side identification text and a client identification text; and according to the conversation voice data of the server side and the corresponding identification text as well as the conversation voice data of the client and the corresponding identification text, carrying out real-time monitoring on the conversation voice of the server side. With the method and system, comprehensive and real-time monitoring of conversation voices of the customer service staff can be realized; and the service quality of the customer service staff is improved.

The invention discloses a customer service voice optimization method, belongs to the technical field of computers, and helps to improve customer service efficiency. The customer service voice optimization method disclosed by the embodiment of the invention comprises the following steps: in response to a received incoming voice of a user, acquiring a voice feature of the user; in response to the received customer service voice of the customer service staff for the incoming call voice, converting the customer service voice based on a preset tone library to obtain the customer service voice matched with the voice feature; and outputting customer service voice matched with the sound features to the user. According to the customer service voice optimization method disclosed by the embodiment of the invention, the customer service voice matched with the voice feature of the user is obtained and output to the user by performing automatic voice conversion on the customer service voice of the customer service staff, so that the auditory experience of the user is improved, and the customer service efficiency is improved.

The invention provides a voice conversion method and apparatus, a device and a storage medium. The method comprises the steps of determining a first low-dimensional vector corresponding to a target voice based on the target voice; separating the first low-dimensional vector to obtain a target identity voice vector; and determining a final conversion voice based on the sample identity voice vector and the target identity voice vector. According to the method, one-to-many voice conversion can be realized without adding an additional conversion model for the system, and meanwhile, a mode of obtaining the target voice by calculating the Mel-frequency cepstrum coefficient MFCC feature offset is adopted, so that a large amount of parallel corpora are not needed as training data, and the difficulty of collecting the training data is reduced; the method can be applied to the customer service platform, the one-to-many voice conversion technology is adopted to increase the types of convertible voices of the customer service system aiming at the problem that the voice conversion function of the customer service platform is single in voice, and the purpose of protecting the privacy of customers is achieved; the system is convenient and simple, has powerful functions, and can effectively and quickly realize voice conversion.