TW202230231A - Judgment system for suitability of talents and implementation method thereof - Google Patents

Abstract

Judgment system for the suitability of talents and implementation method thereof are disclosed. The present invention provides a calculation model based on voiceprint features related to each personality to obtain a personality information from a voiceprint of one of users under test, or obtain a group of the personality information from each of the voiceprint of a group of the users under test. In this way, the judgment system for the suitability of talents and implementation method thereof can immediately obtain the personality and career suggestions of the users, and obtain the compatibility between the personality information of the user and the personality information of the group of the users.

Description

Talent suitability judgment system and method

A system and method for judging suitability between people and jobs, especially a talent suitability judging system and method for understanding personal characteristics from voiceprint features and analyzing occupational suitability through voiceprint recognition technology.

Function refers to the combination of a set of knowledge, skills, behaviors and attitudes. When applied to work, it is used as a reference for checking whether one is competent for a certain task. Among them, behaviors and attitudes are judged based on the more difficult past education and qualifications. , most of which are checked in the form of psychological tests or scales; and psychological tests are divided into two types: verbal and non-verbal. It is relatively easy and fast to implement. However, language tests are relatively standardized and have few controllable variables. They are not suitable for people who have difficulty in language or text recognition, and it is difficult to compare test subjects with different language and cultural backgrounds, and it is easy to make the test subjects difficult. The tester deliberately manipulates the test results, and cannot accurately reflect the behavior and attitude of the subjects; non-verbal tests include various tests that use drawings, tools, and real objects as the test medium. Explain or operate the test medium in real time to reflect the psychological image to the tester. Non-verbal tests are suitable for people with language impairments, and are also suitable for comparing testees with different language and cultural backgrounds, and it is difficult for the testee to deliberately guide the test. Some results, but compared with language tests, have disadvantages such as high implementation cost and long implementation time; in addition, after most tests are administered, they can only briefly understand personality, and cannot be applied to interviews and interviews immediately based on the evaluation results. For example, after understanding the personality of the interviewee, it is impossible to adjust the content of the interview according to the personality characteristics. In addition, job seekers can only understand the company's ecology based on the company's public information and partial information on the Internet, so that when making career planning, Departments often can only choose based on salary and benefits, but cannot know their own and company culture, or whether their personality can get along with the department; based on this, how can we effectively improve the traditional methods of evaluating personality so that It can achieve the advantages of taking into account the efficiency, accuracy and popularization, which is actually required by the current industry.

In view of the above-mentioned problems, the inventors of the present invention research and improve the personality analysis method based on years of experience in related industries; therefore, the main purpose of the present invention is to provide a talent suitability judgment system and method.

In order to achieve the above-mentioned purpose, the system and method for judging human suitability of the present invention includes a server, a voice capture device, and a user-end device. The server is based on voiceprint information and a corresponding character information, With the supervised learning method as the core, a computing model is constructed; then a voiceprint information to be tested is captured through a voice capture device, and the server is used for calculation to obtain its corresponding character, and then the character information is combined with The database in the server is compared, and a character analysis information can be received by the user terminal device, so that the user can quickly and timely understand the character of the person to be tested, as well as suggestions and responses related to the character.

In order to enable your examiners to clearly understand the purpose, technical features and effects of the present invention, the following descriptions are combined with the diagrams for illustration, please refer to.

Please refer to "Fig. 1" and "Fig. 2", which are schematic diagrams (1) and (2) of the present invention. As shown in the figures, the human suitability judgment system 1 of the present invention includes a server 11 , and a voice capture device 12 and a user-end device 13 are connected to the server 11 for information. The functions of each component are illustrated below: (1) The server 11 has an arithmetic processing module 111, and further includes a voiceprint capture module 112, a voiceprint processing module 113, a storage module 114, a data analysis module 115 and an arithmetic processing module 111 Information connection; wherein, the operation processing module 111 is used to run the server 11 and drive the actions of the above modules. The operation processing module 111 has functions such as logical operation, temporary storage of operation results, and storage of execution command positions. It is a central processing unit (Central Processing Unit, CPU), but not limited to this; (2) The voiceprint capturing module 112 can receive a voiceprint information to be tested. The voiceprint information to be tested can be converted based on a voice, or the voiceprint capturing module 112 can capture audio and video files to obtain the voiceprint information to be tested. Measured voiceprint information, the voiceprint information to be measured may be digital or analog voiceprint information such as uncompressed, lossy compression, lossless compression, etc., or a combination thereof, but not limited to this; (3) The voiceprint processing module 113 can perform one or a combination of noise reduction, preprocessing, and feature extraction on the received voiceprint information to be measured, so as to purify the sound to be measured. The pattern information is used by the data analysis module 115. The pre-processing may include Pre-emphasis, Frame, Window, and Endpoint Detection. ), but not limited thereto, the Feature Extraction method may include Linear Predictive Coding Coefficient (LPCC), Mel-frequency Cepstrum Coefficient (MFCC), etc., but not limited to this; (4) The storage module 114 can be used to store electronic data, which can be a solid state hard disk (Solid State Disk or Solid State Drive), a hard disk (Hard Disk Drive, HDD) but is not limited thereto; the data storage module 114 further includes a training database 1141 and a character database 1142, the training database 1141 stores at least voiceprint information and at least one character information corresponding to it; the character database 1142 stores character information Corresponding character analysis information, character analysis information is the analysis and characteristics of character characteristics, such as: behavioral characteristics of character performance, suitable jobs, suitable learning methods, character representatives, reactions to various situations, communication strategies, etc., but not limited thereto; (5) The data analysis module 115 uses artificial intelligence (AI), machine learning (Machine Learning), deep learning (Deep Learning) and other integrated operation modes and their combinations, but not limited to this, it is based on The voiceprint information in the training database 1141 is used as the input data, and the character information is used as the target data. The voiceprint information includes the voiceprint features such as frequency, period, wavelength, vibration, wave pattern, and speech rate in the voiceprint. , but not limited to this, a supervised learning method is used as the core to construct an operation model; and the data analysis module 115 can generate its corresponding character information based on the voiceprint information to be tested, and can also generate the character based on the generated character information, the corresponding character analysis information is retrieved from the character database 1142; (6) The sound capture device 12 is a transducer that can convert sound into electronic signals, such as a microphone, a microphone, and a microphone, but not limited to this. The sound capture device 12 is used to capture sound, And convert the sound into the voiceprint information to be tested; (7) The client device 13 is a device that can receive personality analysis information for reference by the tester or the person to be tested, such as: a computer, a tablet, a smart phone, a monitor, a printer, a projector, a speaker, etc. , but not limited to this.

Please refer to "Fig. 3", which is a flow chart of the implementation of the present invention, and please refer to "Fig. 1" and "Fig. 2" in combination. As shown in the figures, the implementation steps of the talent suitability judgment method of the present invention are as follows : (1) A data analysis module training step S1: Please refer to “Fig. 4” for reference. The data analysis module 115 takes the supervised learning method as the core, constructs an operation model, and is based on the voiceprint in the training database 1141 Information D3 is used as input data, and personality information D4 is used as target data for training. Among them, voiceprint information D3 includes voiceprint features such as frequency, period, wavelength, vibration, wave pattern, and speech speed in the voiceprint, but it does not. As a limitation, the tester can first use various questionnaires and personality test tools to obtain the voiceprint information D3 corresponding to the specific personality information D4. For example, use the Enneagram test on a group of audiences and record their voices together. Each voiceprint information D3 corresponding to each character information D4 of the Enneagram can be obtained, and a training database 1141 can be established based on this, so as to use the training database 1141 to train the data analysis module 115. The above Enneagram test is only for Examples are given, but not limited thereto; (2) A sound recording step S2: Please refer to "Fig. 5", the tester can use the sound capture device 12 to capture a sound D1, wherein the sound D1 can include but not limited to capturing the live and online speech of the test subject The voice, or the voice containing the audio and video files of the subject to be tested, is converted into the voiceprint information D2 to be transmitted to the server 11, or the audio and video files containing the subject to be tested can be directly analyzed by the voiceprint capture module 112, The voiceprint information D2 to be tested is retrieved, and the voiceprint information D2 to be tested is received by the server 11. For example, the voice of the job seeker is recorded with a microphone during the interview, or the job seeker uploads an oral self-introduction audio file before the interview. As a resume, the voiceprint capturing module 112 can retrieve the voiceprint information D2 to be tested from the audio file; (3) Voiceprint processing step S3: the voiceprint processing module 113 performs one or a combination of noise reduction, preprocessing, and feature extraction on the voiceprint information D2 to be measured, and purifies the voiceprint information D2 to be tested. Measure the voiceprint information D2 for analysis and processing by the server 11; (4) Voiceprint analysis step S4: The data analysis module 115 generates its corresponding character information D4 based on the voiceprint information D2 to be tested that has undergone the voiceprint processing step S3, for example: the data analysis module 115 is based on job applicants The voiceprint information D2 to be tested produces the character information D4 as "practice makes perfect"; the voiceprint information D2 to be tested includes voiceprint features such as frequency, period, wavelength, vibration, wave pattern, speech rate, etc. , but not limited to this; (5) A personality analysis step S5: Please refer to "Fig. 6" in conjunction with the data analysis module 115 based on the personality information D4 to filter out a corresponding personality analysis information D5 from a personality database 1142, for example: the data analysis module 115 is a Based on the personality information D4 is "practice makes perfect", the personality database 1142 screened out the characteristics of "patient", "concerned about the integrity and quality of work", and "back office, cook, accountant (division), doctor, need Professional Licensed Work" occupational recommendation; (6) A result output step S6: Please refer to "Fig. 7", the server 11 transmits the character information D4 to a user-end device 13 for the tester or the testee to refer to.

Please refer to "Fig. 8", which is another embodiment (1) of the present invention, and please refer to "Fig. 1" ~ "Fig. 7" in conjunction with the step "result output step S6", the present invention further includes There is a model correction step S7. After the server 11 transmits the character information D4 to a user-end device 13, the user can check whether the character analysis information D5 conforms to his own character through the user-end device 13. If it does not match his own character, then The user can create a correction information D6 through the user terminal device 13, the correction information D6 is the voiceprint information D2 to be tested and the corresponding correct character information D4, and transmit the correction information D6 to the server 11. The server 11 can use the correction information D6 as training data to correct the data analysis module 115 .

Please refer to "Picture 9" and "Picture 10", and please refer to "Picture 1" to "Picture 7" in combination. The storage module 114 further includes at least one group personality database 1143, and the members of each group It is a system and method for judging talent suitability of the present invention. The aforementioned group can be, for example, a group formed by an enterprise, a unit or a department, and a plurality of members. The data analysis module 115 calculates a group of personality information based on the test results. D7, the group personality information D7 is the personality bias of the group, or the personality information that matches the group, the data analysis module 115 can calculate the personality information D3 parsed by the voiceprint analysis step S4, and the group personality information D7. In this way, after the voiceprint analysis step S4, preferably, a group fit step S8 can be performed. The fit information D8 is sent to the user terminal device 13 to execute the result output step S6. In this way, if the job seeker is admitted to multiple companies at the same time and wants to make a final selection, he can refer to the fit information D8 to know the person who is similar to his own personality. Departmental atmosphere and culture, enterprises can also understand the characteristics of their own group composition through group personality information D7.

From the above, it can be seen that the system and method for judging talent suitability of the present invention includes a server, a sound capture device, and a user-end device. The server can be based on a training database and use a supervised learning method. The core is to construct an operation model, the server can generate a personality information based on the voiceprint information to be tested captured by the voice capture device, and generate a personality analysis information by comparing the personality database, which can be used by the tester. The terminal device obtains the personality characteristics of the test subject; the test method using the voiceprint characteristics has the characteristics of fast, accurate, and difficult to fake, and can give the test subject advice and results in a timely manner. The degree of fit between groups, according to this, the talent suitability judgment system and method of the present invention solves the pain point of the conventional personality test tools, and achieves the efficiency, accuracy and popularization of personality evaluation methods required by current industry practice. Purpose.

However, the above descriptions are only preferred embodiments of the present invention, and are not intended to limit the scope of the present invention; anyone familiar with the art can make equal changes and modifications without departing from the spirit and scope of the present invention. , all should be covered within the patent scope of the present invention.

To sum up, the effect of the present invention is that it has the patent requirements of "industrial applicability", "novelty" and "progressiveness" of the invention; the applicant should file an invention patent with the Jun Bureau in accordance with the provisions of the Patent Law Application.

1: Talent suitability judgment system 11: Server 111: Operation processing module 112: Voiceprint Capture Module 113:Voiceprint processing module 115: Data Analysis Module 114: Storage Module 1141: Training database 1142: Personality Database 1143: Group Personality Database 12: Sound capture device 13: User device S1: Data analysis module training steps D1: sound S2: Radio Steps D2: Voiceprint information to be tested S3: Voiceprint processing steps D3: Voiceprint Information S4: Voiceprint Analysis Step D4: Personality Information S5: Steps of Personality Analysis D5: Personality Analysis Information S6: Result output step D6: Correction information S7: Model Correction Step D7: Group Personality Information S8: Group fit step D8: fit information

Figure 1 is a schematic diagram (1) of the composition of the present invention. Figure 2 is a schematic diagram (2) of the composition of the present invention. Fig. 3 is a flow chart of the implementation of the present invention. FIG. 4 is a schematic diagram (1) of the implementation of the present invention. FIG. 5 is a schematic diagram (2) of the implementation of the present invention. Fig. 6 is a schematic diagram (3) of the implementation of the present invention. FIG. 7 is a schematic diagram (4) of the implementation of the present invention. FIG. 8 is another embodiment (1) of the present invention. Fig. 9 is another embodiment (2) of the present invention. Fig. 10 is another embodiment (3) of the present invention.

1: Talent suitability judgment system

11: Server

12: Sound capture device

13: User device

Claims (10)

A talent suitability judgment system, which includes: a sound capturing device for capturing at least one sound and converting it into a voiceprint information to be tested; A server, connected to the voice capture device information, includes an operation processing module, a voiceprint capture module, a voiceprint processing module, a storage module, and a data analysis module and the operation module Handling module information links; the computing processing module is used to run the server; the voice capture device is used to capture the voiceprint information to be tested; The voiceprint processing module is used for purifying the voiceprint information to be tested; The storage module includes a training database and a character database, the training database stores at least voiceprint information and at least one character information, and the character database stores at least one character analysis corresponding to the character information News; The data analysis module is based on the voiceprint information in the training database as input data, the character information as target data, and the supervised learning method as the core to construct an operation model; The data analysis module generates the character information based on the voiceprint information to be tested; The data analysis module generates the character analysis information based on the character information and the character database; and A client device is connected to the server information for receiving the character analysis information. The talent suitability judgment system according to claim 1, wherein the character analysis information includes a behavior characteristic of character performance, a suitable job, a suitable learning method, a character representative, a face to various situations one or a combination of the following responses, or a communication strategy. The talent suitability judgment system according to claim 1, wherein the data analysis module can be modified based on a modification information. The talent suitability judgment system of claim 1, wherein the data analysis module further generates a group of personality information based on the plurality of voiceprint information to be tested. The talent suitability judgment system according to claim 4, wherein the storage module further comprises a group character database storing at least one group character information. The talent suitability judgment system according to claim 4, wherein the data analysis module calculates a fit information based on the group personality information and the personality information. A talent suitability judgment method, which includes: The training steps of the voiceprint classifier: a server constructs a computing model by a supervised learning method based on the voiceprint information and a character information A sound collection step: a sound capture device captures at least one sound, converts it into a voiceprint information to be measured, and transmits it to the server; Voiceprint processing step: the server purifies the voiceprint information to be tested; Voiceprint analysis step: the server generates the character information based on the voiceprint information to be tested; and A character analysis step: the server selects a corresponding character analysis information from a character database based on the character information; and A result output step: the server transmits the character information to a client device. The method for judging human suitability according to claim 7, further comprising a model correction step: the server corrects the operation model based on a correction information. The method for judging the degree of human suitability according to claim 7, wherein the voiceprint analysis step further comprises generating a group of personality information based on a plurality of the voiceprint information to be tested. The method for judging talent suitability according to claim 9, further comprising a group fit step: the server calculates a fit based on the group character information and the character information.

Images