Information consultation service system and consultation method thereof

Abstract

The invention discloses an information consultation service system and a consultation method thereof in the technical field of information consultation, and the system comprises an information collection module which is used for collecting image information, voice information and text information; the voice recognition module is used for recognizing voice information; the image recognition module is used for recognizing image information; the intelligent dialogue module is used for analyzing text information contents, summarizing and guiding users, sorting text results and answering the users; the retrieval analysis module is used for performing retrieval analysis in the database; the manual auditing module is used for auditing the correlation of answers and making a solution plan of the question; and the evaluation feedback module is used for collecting user evaluation and supplementing database optimization retrieval. The system is novel and effective, guides the user to refine the problem through the artificial intelligence dialogue model, meets the use requirements of the old and the disabled through various information collection and response modes, helps the old and the disabled to know the outside world, and facilitates the life of the old and the disabled.

Description

Technical Field

[0001] The invention belongs to the technical field of information consultation, and in particular relates to an information consultation service system and a consultation method thereof. Background Art

[0002] Although the development of information consulting services mainly depends on the rapidly developing information technology environment and the growing demand of enterprises for efficient and customized consulting services, with the popularization of the Internet and information technology today, everyone has their own information consulting needs, from career planning to asking for directions, especially the elderly and disabled who have difficulty going out.

[0003] In today's information consulting service system, demand analysis methods are used to gain an in-depth understanding of users' consulting needs. Through in-depth communication and understanding with users, the system can clarify users' specific problems and goals, thereby providing more accurate consulting services. Intelligent consulting methods use artificial intelligence technologies, such as natural language processing and machine learning, to achieve automated consulting and intelligent recommendations. This method can significantly improve consulting efficiency, reduce labor costs, and provide users with a more convenient and efficient consulting service experience.

[0004] However, services cannot be provided well for these special groups. 1. Elderly and disabled people may find it difficult to adapt to the traditional information consultation service system interface due to limited vision, hearing or operating ability. Many information consultation service systems lack barrier-free design for the elderly and disabled people, such as screen reader support, voice input function, high contrast mode, etc. The lack of these functions may prevent them from using the system effectively and thus from obtaining the required consulting services. 2. The needs of the elderly and disabled people are often special and personalized. However, many information consultation service systems lack customized consulting content for this group and cannot provide solutions that meet their needs. Some information consultation service systems may lack relevant professional knowledge and experience when dealing with consultations from the elderly and disabled people. This may result in them not being accurate or comprehensive enough when answering questions, thus failing to meet the actual needs of users. 3. Elderly and disabled people may find it difficult to accurately understand the consultant's answers due to limited hearing, language or cognitive abilities. This may lead to poor communication and affect the effectiveness of consultation.

[0005] Therefore, it is necessary to propose an information consulting service system and its consulting method that can meet the information consulting needs of the elderly and disabled people, deepen the understanding of user problems, and provide professional answers. Summary of the invention

[0006] In order to solve the above problems, the purpose of the present invention is to provide an information consulting service system and a consulting method thereof, which guides users to refine the pertinence of their questions through an artificial intelligence dialogue model, meets the needs of the elderly and disabled through a variety of information collection and response methods, optimizes the speed and accuracy of consultation answers through answer content scoring and feedback, and helps the elderly and disabled people understand the outside world and facilitates their lives.

[0007] In order to achieve the above-mentioned purpose, the technical scheme of the present invention is as follows: an information consulting service system, including an information collection module, a voice recognition module, an image recognition module, an intelligent dialogue module, a retrieval analysis module, a manual review module and an evaluation feedback module,

[0008] An information collection module is used to collect image information, voice information and text information related to the consultation;

[0009] A speech recognition module, used to recognize the collected speech information as text information;

[0010] An image recognition module, used to recognize the collected image information as text information;

[0011] Intelligent dialogue module, used to analyze text information content, summarize and guide users to ask accurate questions, organize text results and answer users;

[0012] The retrieval analysis module is used to perform retrieval analysis in the database according to the accurate questions raised by the user and obtain the retrieval text results;

[0013] A manual review module is used to review the relevance of information and create problem-solving plans based on search results;

[0014] The evaluation feedback module is used to collect user evaluations, organize search results and add them to the database to optimize search and analysis efficiency.

[0015] The principle of the basic scheme is: the information consulting service system of the present invention comprehensively collects consulting information provided by users through information collection module, voice recognition module and image recognition module. The intelligent dialogue module uses artificial intelligence dialogue model to analyze user input, guide users to clarify questions, and generate preliminary answers. The retrieval analysis module retrieves relevant information in the database to provide data support for the answer. The manual review module reviews and optimizes the retrieval results to ensure the quality and accuracy of the answer. The evaluation feedback module continuously optimizes system performance and database content through user evaluation.

[0016] The beneficial effects of the basic solution are: 1. Through the deep natural language processing technology of the intelligent dialogue module, this system can accurately capture and understand the specific needs and problem expressions of the elderly and disabled people, and then guide users to refine their questions and make the consultation content more specific and clear. This meticulous guidance not only improves the personalization level of consulting services, but also ensures that the answers can get to the heart of the problem and effectively meet the actual needs of users.

[0017] 2. Considering the various obstacles that the elderly and disabled people may encounter when using traditional consulting services, this system has specially designed voice recognition modules and image recognition modules. The use of these modules allows users to easily express their consulting needs through voice or images even when it is inconvenient to use text input, greatly improving the convenience and accessibility of information exchange.

[0018] 3. Through the dual guarantee of the search and analysis module and the manual review module, this system can provide users with accurate answers based on authoritative databases and professional knowledge bases. The addition of manual review further ensures the accuracy and practicality of the answers, avoiding the inconvenience caused to users by misleading information.

[0019] 4. As an important part of this system, the evaluation and feedback module not only collects direct feedback from users on consulting services, but also provides valuable references for subsequent optimization of the system by sorting and analyzing these data. This continuous iteration and optimization mechanism enables the system to keep up with changes in user needs and continuously improve service quality and user experience.

[0020] 5. The design and implementation of this system fully reflects the care and respect for vulnerable groups such as the elderly and the disabled. By providing convenient and barrier-free consulting services, it not only helps them better understand external information, but also promotes the overall inclusiveness of society and the construction of a barrier-free environment.

[0021] Furthermore, the information collection module includes a camera for collecting image information, a recorder for collecting voice information, and an input device for inputting text.

[0022] The beneficial effects of the basic solution are: 1. By integrating a camera, a recorder and a text inputter, the information collection module can simultaneously process three types of information: images, voice and text. This comprehensive and diverse information collection method not only meets the consultation needs of the elderly and disabled in different situations, but also provides them with a variety of ways to express consultation questions, greatly improving the flexibility and convenience of information exchange.

[0023] 2. For the elderly and people with disabilities, traditional text input methods may be difficult to operate. The camera and recorder in the information collection module provide them with a more intuitive and convenient input method. Users can complete the input of consulting information through simple voice commands or image capture without cumbersome keyboard operations, which lowers the threshold for use and improves accessibility.

[0024] 3. The images and voice information collected by the camera and recorder can be converted into text information through advanced image recognition and voice recognition technology, providing accurate data support for the subsequent intelligent dialogue module and retrieval analysis module. This automated information processing method not only improves the accuracy of information processing, but also significantly improves the response speed and efficiency of consulting services.

[0025] 4. As the basic component of the system, the comprehensive and diverse information collection capabilities of the information collection module provide a solid foundation for the subsequent functional expansion and upgrading of the system. With the continuous advancement of technology and the changing needs of users, the system can continuously optimize and improve various functions based on the data support of the information collection module to provide more efficient and convenient consulting services.

[0026] Furthermore, the speech recognition module includes an acoustic model unit and a decoding analysis unit.

[0027] Acoustic model unit, used to extract speech acoustic features and convert them into phoneme sequences, including conversion of characteristic phonemes for multiple dialects;

[0028] The decoding and analysis unit is used to obtain a language recognition text result based on the language model according to the extracted phoneme sequence.

[0029] The beneficial effects of the basic solution are as follows: 1. The acoustic model unit can accurately extract the acoustic features in the speech and convert them into a phoneme sequence. The unit is specially optimized for the characteristic phonemes of multiple dialects, effectively improving the recognition ability of dialect speech. This design not only meets the speech recognition needs of users in different regions, but also enhances the generalization ability of the system, making the speech recognition results more accurate and reliable.

[0030] 2. The decoding analysis unit can quickly obtain the text results of language recognition based on the extracted phoneme sequence and combined with the advanced language model. In this process, the efficiency of the decoding algorithm and the accuracy of the language model jointly ensure the real-time and robustness of speech recognition. Even in a noisy environment, the system can maintain a high recognition rate and provide users with a smooth voice interaction experience.

[0031] 3. The voice recognition module allows users to easily express their consulting needs through voice without having to manually input text. The convenience of this input method greatly reduces the user's operating threshold, especially for the elderly and people with disabilities. At the same time, the naturalness and intuitiveness of voice interaction also improves users' satisfaction and trust in consulting services.

[0032] 4. As an important part of the system, the voice recognition module is closely integrated with other modules (such as information collection module, intelligent dialogue module, etc.), together forming a complete information consulting service process. This integrated design not only improves the overall performance and efficiency of the system, but also promotes the collaborative work between modules, providing users with more comprehensive and efficient consulting services.

[0033] Furthermore, the image recognition module includes an image feature extraction unit, a text reading unit and a video action classification unit.

[0034] An image feature extraction unit is used to extract features of objects and people in the image and generate text results;

[0035] A text reading unit, used to read text information in an image and generate a text result;

[0036] The video action classification unit is used to classify and identify the motion patterns of objects and people in the video and generate text results.

[0037] The beneficial effects of the basic solution are: 1. The image feature extraction unit can accurately identify the features of objects and people in the image and convert them into text results. This high-precision feature extraction capability not only improves the accuracy of information extraction, but also significantly improves processing efficiency.

[0038] 2. Through the processing of the image recognition module, the complex information in the image and video is converted into text results that are easy to understand and analyze. This conversion not only makes the information more intuitive and clear, but also facilitates further processing and mining by the subsequent intelligent dialogue module and retrieval analysis module.

[0039] 3. The application of the image recognition module is not limited to traditional text consultation, but can also be extended to the field of image and video consultation. For example, users can consult relevant issues by uploading pictures or videos, and the system can automatically identify and extract key information from them to provide users with more accurate and personalized consulting services. This broadened application scenario not only enriches the form and content of consulting services, but also improves the practicality and flexibility of the system.

[0040] 4. The image recognition module enables users to provide consulting information in a more intuitive and convenient way. Users do not need to manually enter complex descriptions or explanations, but only need to upload relevant images or videos. The convenience of this input method greatly reduces the user's operating difficulty and time cost, and improves user experience and satisfaction.

[0041] Furthermore, the intelligent dialogue module includes an artificial intelligence dialogue model based on natural language processing technology, a text display unit and a voice answer unit.

[0042] The artificial intelligence dialogue model is used to analyze questions raised by users and ask targeted questions to guide users to refine their questions. It is also used to organize and summarize the text results obtained through recognition and retrieval to answer user questions.

[0043] A text display unit, used to display the content of communication and answers;

[0044] Voice answer unit, used to read out the communication and answer content.

[0045] The beneficial effects of the basic solution are: 1. The artificial intelligence dialogue model can accurately understand the questions raised by users, and guide users to refine questions through targeted counter-questions, making the dialogue process more natural and smooth. This guidance method not only helps the system to grasp user needs more accurately, but also improves the efficiency and effectiveness of the dialogue. At the same time, the collaborative work of the text display unit and the voice answer unit enables the dialogue content to be presented in a variety of forms, further enhancing the intuitiveness and convenience of the interaction.

[0046] 2. The AI ​​dialogue model not only has powerful natural language processing capabilities, but can also organize and summarize the text results obtained through identification and retrieval based on user questions to provide users with accurate and comprehensive answers. This capability not only improves the efficiency and accuracy of information retrieval, but also ensures the authority and reliability of the answer content.

[0047] 3. The application of the intelligent dialogue module is not limited to traditional text consultation, but can also be extended to various scenarios such as voice consultation and image consultation. Through the voice answer unit, the system can read the communication and answer content aloud, so that users such as the elderly and the disabled who are not convenient to read text can also easily enjoy consulting services. This broadened application scenario and audience range not only improves the practicality and popularity of the system, but also reflects the care and respect for vulnerable groups.

[0048] Furthermore, the retrieval analysis module includes a database, a vocabulary retrieval unit and a logic retrieval unit.

[0049] Databases are used to provide channels for retrieval work and continuously accumulate new feedback data;

[0050] A vocabulary retrieval unit is used to retrieve data based on keywords in user questions and obtain relevant data and relevance;

[0051] The logical retrieval unit is used to logically classify the relevant data according to the degree of association to obtain the retrieval text results.

[0052] The beneficial effects of the basic solution are: 1. The vocabulary search unit can accurately capture the keywords in the user's questions and quickly search for relevant data in the database. This keyword-based search method not only improves the accuracy of information retrieval, but also significantly improves the search efficiency. At the same time, the logical search unit can logically classify the relevant data according to the degree of relevance, further filter out the information that best meets the user's needs, and ensure the accuracy and pertinence of the search results.

[0053] 2. The retrieval analysis module integrates advanced algorithms and models to achieve intelligent and automated information processing. From keyword extraction, data retrieval to logical classification, the entire process does not require manual intervention, which greatly reduces the burden on staff and improves work efficiency. At the same time, this intelligent processing method also enables the system to continuously learn and optimize, and continuously improve retrieval performance.

[0054] 3. As the foundation of the retrieval and analysis module, the database continuously accumulates new feedback data, providing rich channels and resources for retrieval work. This continuous data accumulation not only enriches the dimensions of information retrieval, but also enables the system to deeply explore the correlation and regularity between data, providing users with more comprehensive and in-depth information services.

[0055] Furthermore, the manual review module includes a relevance spot check unit and a manual planning unit.

[0056] The relevance spot check unit is used for staff to spot check the content of the answers to questions and verify the relevance of the questions and answers;

[0057] The human curation unit is used for staff to compile creative answers to complex questions based on the search text results.

[0058] The beneficial effects of the basic scheme are: 1. The relevance sampling unit effectively verifies the relevance between questions and answers through staff sampling, ensuring the accuracy and pertinence of information. This manual review method can timely discover and correct possible deviations or misunderstandings that may occur during the automatic processing of the machine, thereby improving the overall quality of information services.

[0059] 2. The manual curation unit allows staff to compile creative answers to complex questions based on the search text results. This personalized answer method not only meets the user's needs for in-depth information and unique insights, but also enhances the value and attractiveness of information services. Through manual curation, the system can provide richer, more diverse and insightful answers, enhancing user satisfaction and trust.

[0060] 3. The existence of the manual review module provides a reliable backing for information consulting services. Through strict review and planning by staff, the system can ensure that the information provided is reliable and the content is accurate, thereby maintaining the authority and credibility of the information service. This is of great significance for building user trust and enhancing brand image.

[0061] 4. Although the manual review module adds manual intervention, the entire workflow can be optimized and work efficiency can be improved through reasonable process design and division of labor. For example, relevance spot checks can be conducted on some answers to reduce workload; manual planning can focus on complex or high-value issues to ensure effective use of resources.

[0062] 5. The implementation of the manual review module requires staff to have high professional quality and review capabilities. By participating in the review and planning work, team members can continuously learn and accumulate professional knowledge and improve their personal skills. At the same time, communication and collaboration within the team can also promote knowledge sharing and experience inheritance, providing strong support for the continuous optimization and development of the information consulting service system.

[0063] Furthermore, the evaluation feedback module includes an evaluation collection unit and a result feedback unit.

[0064] An evaluation collection unit, used to obtain the user's evaluation of the answer, including ratings and text evaluations;

[0065] The result feedback unit is used to feed back the obtained text results and answer contents to the database and prioritize the answer contents according to the scores.

[0066] The beneficial effects of the basic solution are: 1. The evaluation collection unit can directly obtain the user's evaluation of the answer, including specific scores and text evaluations. These feedbacks are a direct reflection of the user's true feelings, which helps the system to accurately understand the user's needs and expectations, thereby improving the service content and the quality of the answer in a targeted manner. Through continuous optimization, the system can more accurately meet user needs and improve user satisfaction.

[0067] 2. The result feedback unit feeds back the obtained text results and answer content to the database, and prioritizes the answer content according to the user's rating. This ranking method based on user feedback enables the system to prioritize high-quality and user-satisfied answers, improving the efficiency of information retrieval and presentation.

[0068] 3. The evaluation and feedback module not only provides a mechanism for collecting user feedback, but also provides an important basis for the system's self-learning and iteration. By analyzing the user's ratings and textual evaluations, the system can discover problems and deficiencies in the answers, and then adjust the algorithm model and optimize the answer strategy. This continuous learning and iteration process enables the system to continuously adapt to changes in user needs and provide more intelligent and personalized services.

[0069] 4. When users encounter similar questions, the system can quickly retrieve the evaluation data of previous users on similar questions. These evaluation data not only include direct feedback from users on the quality of answers (such as ratings), but also include specific text evaluations, such as which answers are popular with users and which need to be improved. The system can use this information to provide users with more accurate and personalized answers, thereby improving the user experience.

[0070] 5. Through the evaluation collection unit, the system can accumulate a large amount of user answer preference data for similar questions. This data helps the system identify which answer templates or strategies are more popular with users, so that when encountering similar questions, it can automatically select or generate answers that are more in line with user expectations. This optimization strategy not only improves the accuracy of answers, but also reduces the need for manual intervention and improves work efficiency.

[0071] An information consulting service method, characterized in that it comprises the following steps:

[0072] Step 1: Ask a question. The user asks a language or text question using a recorder or input device.

[0073] Step 2: Question analysis and guidance: the intelligent dialogue model analyzes the question text and guides the user to refine the question;

[0074] Step 3: result identification or retrieval: according to the refined targeted questions, the retrieval analysis module obtains the retrieval text results from the database;

[0075] Step 4: Organize answers. The intelligent dialogue module answers the user's questions based on the search text results, or the staff in the manual review module compiles answers based on the search text results.

[0076] Step 5: User evaluation: after receiving the answer, the user will rate and evaluate the answer in the evaluation collection unit according to their satisfaction;

[0077] Step six, feedback optimization, the result feedback unit feeds back the existing text results and answer content to the database, and arranges the priority of the answer content according to the score.

[0078] Furthermore, the result recognition in step three is when the user requires the recognition of actual objects, people and scenes, the information collection module, speech recognition module and image recognition module are used to collect and recognize the target again, and the recognition text result is obtained for the next step. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 Schematic diagram of an information consulting service system in an embodiment of the present invention.

[0080] Figure 2 Schematic diagram of an information collection module of an information consulting service system in an embodiment of the present invention.

[0081] Figure 3 Schematic diagram of a speech recognition module of an information consulting service system in an embodiment of the present invention.

[0082] Figure 4 Schematic diagram of an image recognition module of an information consulting service system in an embodiment of the present invention.

[0083] Figure 5 Schematic diagram of an intelligent dialogue module of an information consulting service system in an embodiment of the present invention.

[0084] Figure 6 It is a schematic diagram of a retrieval and analysis module of an information consulting service system in an embodiment of the present invention.

[0085] Figure 7 Schematic diagram of a manual review module of an information consulting service system in an embodiment of the present invention.

[0086] Figure 8 Schematic diagram of an evaluation feedback module of an information consulting service system in an embodiment of the present invention.

[0087] Fig. 9 Schematic diagram of the information consulting service method in an embodiment of the present invention. DETAILED DESCRIPTION

[0088] The following is further described in detail through specific implementation methods:

[0089] Example 1

[0090] Basically as attached Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 , Figure 8 and Fig. 9 As shown: An information consulting service system, characterized by: including an information collection module, a voice recognition module, an image recognition module, an intelligent dialogue module, a retrieval analysis module, a manual review module and an evaluation feedback module,

[0091] The information collection module is used to collect image information, voice information and text information related to the consultation. The information collection module includes a camera for collecting image information, a recorder for collecting voice information and an input device for inputting text.

[0092] The speech recognition module is used to recognize the collected speech information as text information. The speech recognition module includes an acoustic model unit, which is used to extract speech acoustic features and convert them into phoneme sequences, including conversion of characteristic phonemes for multiple dialects; a decoding analysis unit, which is used to obtain language recognition text results based on the language model according to the extracted phoneme sequence.

[0093] The image recognition module is used to identify the collected image information as text information. The image recognition module includes an image feature extraction unit, which is used to extract the features of objects and people in the image and generate text results; a text reading unit, which is used to read the text information in the image and generate text results; and a video action classification unit, which is used to classify and identify the motion patterns of objects and people in the video and generate text results.

[0094] The intelligent dialogue module is used to analyze the content of text information, summarize and guide users to ask accurate questions, organize the text results and answer users. The intelligent dialogue module includes an artificial intelligence dialogue model based on natural language processing technology, which is used to analyze the questions raised by users and ask targeted questions to guide users to refine their questions, and to organize and summarize the text results obtained by recognition and retrieval to answer user questions; a text display unit is used to display the communication and answer content; and a voice answer unit is used to read aloud the communication and answer content.

[0095] The retrieval analysis module is used to perform retrieval analysis in the database according to the accurate questions raised by the user to obtain the retrieval text results. The retrieval analysis module includes a database, which is used to provide channels for retrieval work and continuously accumulate new feedback data; a vocabulary retrieval unit, which is used to retrieve data according to keywords in the user's questions to obtain relevant data and relevance; a logical retrieval unit, which is used to logically classify relevant data according to the relevance to obtain retrieval text results.

[0096] The manual review module is used to review the relevance of information based on the search results and to develop solutions to problems. The manual review module includes a relevance spot check unit, which is used for staff to spot check the content of answers to questions and verify the relevance of questions and answers; and a manual planning unit, which is used for staff to compile creative answers to complex questions based on the search text results.

[0097] The evaluation feedback module is used to collect user evaluations, organize search results and supplement them into the database to optimize search and analysis efficiency. The evaluation feedback module includes an evaluation collection unit, which is used to obtain user evaluations of answers, including scores and text evaluations; a result feedback unit, which is used to feed back the obtained text results and answer content to the database, and arrange the priority of the answer content according to the score.

[0098] The specific implementation process is as follows: Since a large number of elderly people and disabled people also have information consultation needs, when they use this system, they can first input question text through the input device, or input voice to convert into question text (such as patients with speech disorders or people with physical disabilities). Taking into account regional language differences, the acoustic model unit of the speech recognition module has built-in multiple dialect phoneme models for recognition and conversion. After decoding, the most accurate question recognition text is obtained to help the elderly and disabled people flexibly express their needs.

[0099] Based on the question recognition text, the artificial intelligence dialogue model analyzes the content and keywords of the question text, and uses the text display unit and the voice answer unit to further respond and ask questions in return, guiding the user to refine the questions they ask. For example, if the user's original question is "How do I pass the exam?", the artificial intelligence dialogue model will guide the user to communicate through questions, supplement information such as exam subjects, knowledge that the user believes to be missing, and reference books, enrich the search conditions, and make the final consultation answer results more accurate. At the same time, human-computer dialogue interaction can also be achieved through a variety of methods such as text display or voice broadcast, further facilitating consultation for the elderly and disabled (especially those with visual impairments and hearing impairments).

[0100] After obtaining accurate questions through guidance and refinement, the retrieval analysis module will retrieve the answers to the accurate questions in the database, and obtain relevant retrieval text results from keywords and question logic (such as what and why). The obtained retrieval text results will be summarized and sorted again by the artificial intelligence dialogue model to become a well-structured and easy-to-understand text form, and respond to users through flexible answering methods. These answer text results may have a weak correlation with the questions asked by users. The correlation sampling unit in this system is an insurance mechanism for manual intervention. Through the form of sampling, it not only reduces the consumption of human resources, but also allows staff to understand the working status of the system and the effect of consulting services, helping staff to better optimize the system operation efficiency and retrieval accuracy. In addition, when the questions asked by users are more complex, such as "make a study plan", the system is obviously not enough to rely on retrieval and artificial intelligence dialogue models to obtain a complete and appropriate study plan. The manual planning unit is specifically used to arrange staff to compile creative answers to these complex questions based on the retrieval text results.

[0101] After the user receives the answer, he / she will score it according to his / her satisfaction level and can also provide textual suggestions for improving the answer. These evaluations will be added to the database along with the answer text and the result feedback unit, and the answer text will be prioritized according to the score. On the one hand, when a user asks the same question again, the retrieval and analysis module can directly call the existing high-priority answers to speed up the operation efficiency of the system. On the other hand, the priority ranking can reveal the user's attention aspects for information consultation, making it easier for the staff to summarize and optimize the retrieval and analysis module to improve the effectiveness of information consultation answers.

[0102] Example 2

[0103] The difference from the above embodiment is that, as shown in the attached Figure 1 , Figure 2 , Figure 3 , Figure 4 , Figure 5 and Fig. 9 As shown: The main reason why the elderly and disabled people have difficulty communicating with the outside world is often the decline or damage of their mobility or perception ability, so they also need a system that can help solve this problem, especially the visually impaired, hearing impaired, etc.

[0104] When they ask questions about perceiving the environment (for example, where are we now, what is written in the text in front of me, what is on the table and what the person in front is doing), the intelligent dialogue module recognizes the immediate perception questions raised by the users, and the system can collect the image information and voice information that need to be perceived through the information collection module. The image feature extraction unit extracts relevant object and character information (such as object color and item type, character gender and character action), the text reading unit reads the text, the video motion classification unit recognizes the movement patterns of objects and characters (for example, someone is running in front, someone falls and lies on the ground, and many people are fighting together), and the voice recognition module analyzes the voice text information (such as the content of the speech and the simple source of the sound). The intelligent dialogue module converts all this information into text information, and displays or reads aloud to answer the users, helping users perceive changes in the surrounding environment, care for the elderly and the disabled, and enhance their sense of security.

[0105] Example 3

[0106] The difference from the above embodiment is that, as shown in the attached Fig. 9 As shown: An information consulting service method, comprising the following steps:

[0107] Step 1, asking questions, the user uses a recorder or input device to ask language or text questions.

[0108] Step 2: Question analysis and guidance: the intelligent dialogue model analyzes the question text and guides the user to refine the question.

[0109] Step three, result recognition or retrieval, based on the refined targeted questions, the retrieval analysis module obtains the retrieval text results from the database; result recognition is when the user requires the recognition of actual objects, people and scenes, the information collection module, voice recognition module and image recognition module are used to collect and recognize the target again to obtain the recognition text results.

[0110] Step 4: Organize answers. The intelligent dialogue module answers the user's questions based on the search text results, or the staff in the manual review module compiles answers based on the search text results.

[0111] Step 5: User evaluation. After receiving the answer, the user will rate and evaluate the answer in the evaluation collection unit based on his or her satisfaction.

[0112] Step six, feedback optimization, the result feedback unit feeds back the existing text results and answer content to the database, and arranges the priority of the answer content according to the score.

[0113] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device 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 device.

[0114] The above is only an embodiment of the present invention. The common sense such as the known specific structure and characteristics in the scheme is not described in detail here. The ordinary technicians in the relevant field know all the common technical knowledge in the technical field of the invention before the application date or priority date, can obtain all the existing technologies in the field, and have the ability to apply the conventional experimental means before that date. The ordinary technicians in the relevant field can improve and implement this scheme in combination with their own abilities under the enlightenment given by this application. Some typical known structures or known methods should not become obstacles for ordinary technicians in the relevant field to implement this application. It should be pointed out that for those skilled in the art, without departing from the structure of the present invention, several deformations and improvements can be made, which should also be regarded as the protection scope of the present invention, which will not affect the effect of the implementation of the present invention and the practicality of the patent. The protection scope required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. An information consulting service system, characterized in that: It includes information collection module, voice recognition module, image recognition module, intelligent dialogue module, retrieval analysis module, manual review module and evaluation feedback module. An information collection module is used to collect image information, voice information and text information related to the consultation; A speech recognition module, used to recognize the collected speech information as text information; An image recognition module, used to recognize the collected image information as text information; Intelligent dialogue module, used to analyze text information content, summarize and guide users to ask accurate questions, organize text results and answer users; The retrieval analysis module is used to perform retrieval analysis in the database according to the accurate questions raised by the user and obtain the retrieval text results; A manual review module is used to review the relevance of information and create problem-solving plans based on search results; The evaluation feedback module is used to collect user evaluations, organize search results and add them to the database to optimize search and analysis efficiency.

2. The information consulting service system according to claim 1, characterized in that: The information collection module includes a camera for collecting image information, a recorder for collecting voice information, and an input device for inputting text.

3. The information consulting service system according to claim 1, characterized in that: The speech recognition module includes an acoustic model unit and a decoding analysis unit. Acoustic model unit, used to extract speech acoustic features and convert them into phoneme sequences, including conversion of characteristic phonemes for multiple dialects; The decoding and analysis unit is used to obtain a language recognition text result based on the language model according to the extracted phoneme sequence.

4. The information consulting service system according to claim 1, characterized in that: The image recognition module includes an image feature extraction unit, a text reading unit, and a video action classification unit. An image feature extraction unit is used to extract features of objects and people in the image and generate text results; A text reading unit, used to read text information in an image and generate a text result; The video action classification unit is used to classify and identify the motion patterns of objects and people in the video and generate text results.

5. The information consulting service system according to claim 1, characterized in that: The intelligent dialogue module includes an artificial intelligence dialogue model based on natural language processing technology, a text display unit and a voice answer unit. The artificial intelligence dialogue model is used to analyze questions raised by users and ask targeted questions to guide users to refine their questions. It is also used to organize and summarize the text results obtained through recognition and retrieval to answer user questions. A text display unit, used to display the content of communication and answers; Voice answer unit, used to read out the communication and answer content.

6. The information consulting service system according to claim 1, characterized in that: The retrieval analysis module includes a database, a vocabulary retrieval unit and a logical retrieval unit. Databases are used to provide channels for retrieval work and continuously accumulate new feedback data; A vocabulary retrieval unit is used to retrieve data based on keywords in user questions and obtain relevant data and relevance; The logical retrieval unit is used to logically classify the relevant data according to the degree of association to obtain the retrieval text results.

7. The information consulting service system according to claim 1, characterized in that: The manual review module includes a relevance spot check unit and a manual planning unit. The relevance spot check unit is used for staff to spot check the content of the answers to questions and verify the relevance of the questions and answers; The human curation unit is used for staff to compile creative answers to complex questions based on the search text results.

8. The information consulting service system according to claim 1, characterized in that: The evaluation feedback module includes an evaluation collection unit and a result feedback unit. An evaluation collection unit, used to obtain the user's evaluation of the answer, including ratings and text evaluations; The result feedback unit is used to feed back the obtained text results and answer contents to the database and prioritize the answer contents according to the scores.

9. An information consulting service method, characterized in that: The following steps are involved: Step 1: Ask a question. The user asks a language or text question using a recorder or input device. Step 2: Question analysis and guidance: the intelligent dialogue model analyzes the question text and guides the user to refine the question; Step 3: result identification or retrieval: according to the refined targeted questions, the retrieval analysis module obtains the retrieval text results from the database; Step 4: Organize answers. The intelligent dialogue module answers the user's questions based on the search text results, or the staff in the manual review module compiles answers based on the search text results. Step 5: User evaluation: after receiving the answer, the user will rate and evaluate the answer in the evaluation collection unit according to their satisfaction; Step six, feedback optimization, the result feedback unit feeds back the existing text results and answer content to the database, and arranges the priority of the answer content according to the score.

10. The information consulting service method according to claim 9, characterized in that: The result recognition in step three is when the user requires the recognition of actual objects, people and scenes. The information collection module, speech recognition module and image recognition module are used to collect and recognize the target again to obtain the recognition text result for the next step.

Images