Biological recognition and acquisition device and conference summary equipment

Abstract

The utility model provides a biological recognition and acquisition device and conference summary equipment. The device comprises a biological information acquisition mechanism, a movable supporting mechanism and a data processor, a face acquisition module, a first voice acquisition module and a second voice acquisition module in the biological information acquisition mechanism are electrically connected with a data processor; a connecting rod in the movable supporting mechanism is connected with the biological information collecting mechanism, and a rotating joint in the movable supporting mechanism is rotationally connected with the connecting rod and the telescopic rod. The face acquisition module determines identity information of a speaker according to a face recognition result, the first voice acquisition module and the second voice acquisition module are arranged on the two sides of the face acquisition module, and the accuracy of sound source positioning and voice recognition is improved; the rotary joint drives the biological information acquisition mechanism to perform angle adjustment, and the telescopic rod drives the biological information acquisition mechanism to perform height adjustment, so that the biological information acquisition mechanism directly faces the face of the speaker, and the accuracy of recognition of facial information and voice information of the speaker is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the technical field of conference record keeping equipment, and in particular, to a biometric identification and collection device and a conference record keeping equipment. Background Art

[0002] Meeting minutes equipment is required in meeting scenarios to record and summarize the meeting content. The meeting minutes equipment includes a voice recognition sensor, which performs voiceprint recognition and voice recognition on the speaker through the voice recognition sensor, and generates meeting minutes based on the content recognized by the voice recognition sensor.

[0003] However, there are multiple speakers during a meeting, and the location of the conference record device is fixed and easily disturbed by environmental factors, so it is impossible to accurately identify the voiceprints and speech content of speakers in different locations. For example, when the angle of the conference record device is not facing the face of the current speaker, the voice recognition sensor installed on the conference record device will not only collect the speaker's voice information, but also collect interference information such as noise in the environment, reducing the accuracy of voice and voiceprint recognition. Utility Model Content

[0004] The present application provides a biometric identification and collection device and a conference minutes device to improve the accuracy of the conference minutes device in recognizing the voices and voiceprints of different speakers.

[0005] In a first aspect, the present application provides a biometric identification and collection device, including: a biometric information collection mechanism, a movable support mechanism, and a data processor;

[0006] The biometric information collection mechanism includes a face collection module, a first voice collection module, a second voice collection module, a first rotating shaft and a second rotating shaft, and the face collection module, the first voice collection module, and the second voice collection module are electrically connected to the data processor;

[0007] The face acquisition module, the first voice acquisition module and the second voice acquisition module are rectangular parallelepiped structures, and the first rotating shaft and the second rotating shaft are cylindrical structures;

[0008] The first rotating shaft is arranged at one side of the face acquisition module, and the first rotating shaft is rotatably connected to the face acquisition module and the first voice acquisition module, and the second rotating shaft is arranged at the other side of the face acquisition module, and the second rotating shaft is rotatably connected to the face acquisition module and the second voice acquisition module;

[0009] The movable support mechanism includes a connecting rod, a rotating joint and a telescopic rod, and the movable support mechanism is electrically connected to the data processor;

[0010] The connecting rod is a cylindrical structure, the rotating joint is a spherical structure, the telescopic rod includes a first sliding rod and a second sliding rod, the first sliding rod and the second sliding rod are cylindrical structures, and the first sliding rod is slidably connected to the second sliding rod;

[0011] The connecting rod is connected with the biological information collection mechanism, and the rotating joint rotates to connect the connecting rod and the telescopic rod.

[0012] In the above-mentioned biometric identification and collection device, the face collection module in the biometric information collection mechanism determines the identity information of the speaker based on face recognition, and the first voice collection module and the second voice collection module are respectively arranged on both sides of the face collection module to improve the accuracy of sound source positioning and voice recognition; the first rotating shaft and the second rotating shaft drive the first voice collection module and the second voice collection module to rotate, and control the expansion and closing states of the first voice collection module and the second voice collection module; the rotating joint in the movable support mechanism drives the connecting rod to rotate, so that the connecting rod drives the biometric information collection mechanism to adjust the angle, and the telescopic rod drives the biometric information collection mechanism to adjust the height. Through the angle and height adjustment, the biometric information collection mechanism can face the speaker's face, thereby improving the accuracy of recognizing the speaker's facial information and voice information.

[0013] In an implementation manner of the first aspect, the face acquisition module includes a first surface and a face recognition sensor, the first surface is rectangular, the face recognition sensor is a nested structure, and the face recognition sensor is arranged on the first surface;

[0014] The first voice collection module includes a second surface and a first voice recognition sensor, the second surface is rectangular, a hollow structure is provided on the housing of the first voice recognition sensor, and the first voice recognition sensor is arranged on the second surface;

[0015] The second voice collection module includes a third surface and a second voice recognition sensor. The third surface is rectangular. A hollow structure is provided on the shell of the second voice recognition sensor. The second voice recognition sensor is arranged on the third surface.

[0016] In the above-mentioned biometric identification and collection device, a face recognition sensor is installed on the face collection module, a first voice recognition sensor is installed on the first voice collection module, and a second voice recognition sensor is installed on the second voice collection module, so that the face recognition sensor, the first voice recognition sensor and the second voice recognition sensor do not interfere with each other, thereby achieving the purpose of simultaneously identifying the speaker's facial information, voiceprint information and language information.

[0017] In an implementation manner of the first aspect, the face recognition sensor includes a camera module;

[0018] The data processor comprises a first input interface;

[0019] The camera module is electrically connected to the first input interface.

[0020] In the above-mentioned biometric identification and acquisition device, the camera module has the functions of capturing real-time video, identifying and locating targets, tracking target movement, performing three-dimensional space mapping, and scene analysis and understanding, etc., to provide support for accurate and efficient face recognition functions. The camera module is electrically connected to the first input interface of the data processor, and the acquired facial information of the speaker is input into the data processor for processing.

[0021] In an implementation of the first aspect, the first speech recognition sensor includes a first microphone array;

[0022] The data processor also includes a second input interface;

[0023] The first microphone array is electrically connected to the second input interface.

[0024] In an implementation manner of the first aspect, the second speech recognition sensor includes a second microphone array;

[0025] The data processor also includes a third input interface;

[0026] The second microphone array is electrically connected to the third input interface.

[0027] In the above-mentioned biometric identification and collection device, the first microphone array and the second microphone array are used to enhance the target voice, suppress background noise and interference, extract and separate signals, thereby improving the accuracy and reliability of voice recognition. The first microphone array and the second microphone array are electrically connected to the second input interface and the third input interface of the data processor respectively, and the optimized voice information is input into the data processor for processing.

[0028] In an implementation manner of the first aspect, the movable support mechanism further includes a first drive motor, a second drive motor, a reducer, and a lead screw, wherein a first output shaft of the first drive motor is connected to the rotary joint through the reducer, and a second output shaft of the second drive motor is connected to the telescopic rod through the lead screw;

[0029] The data processor also includes a first output interface;

[0030] The first drive motor and the second drive motor are electrically connected to the first output interface.

[0031] In the above-mentioned biometric identification and collection device, the first drive motor and the second drive motor are both electrically connected to the first output interface of the data processor. After the data processor processes the received facial and voice data, it inputs an adjustment signal to the first drive motor and the second drive motor. After receiving the adjustment signal, the first output shaft in the first drive motor transmits the rotation to the rotary joint through the reducer, driving the rotary joint to rotate for angle adjustment; the second output shaft in the second drive motor converts the rotation into linear motion through the lead screw and transmits it to the telescopic rod, driving the telescopic rod to adjust the height. In this way, the angle and height of the biometric information collection mechanism are adjusted, so that the face collection module, the first voice collection module and the second voice collection module in the biometric information collection mechanism face the speaker's face, thereby improving the accuracy of information collection.

[0032] In an implementation manner of the first aspect, the data processor further includes a second output interface;

[0033] The second output interface is electrically connected to an external device.

[0034] In the above-mentioned biometric identification and collection device, the data processor processes the collected facial and voice data with high accuracy to generate meeting minutes, and inputs them to the external device through the second output interface, which is then stored or displayed by the external device.

[0035] In an implementation manner of the first aspect, the length of the second surface is 1 / 2 of the length of the first surface, and the height of the second surface is equal to the height of the first surface;

[0036] The length of the third surface is 1 / 2 of the length of the first surface, and the height of the third surface is equal to the height of the first surface.

[0037] In the above-mentioned biometric identification and collection device, the first voice collection module and the second voice collection module are the same in size and shape, and their lengths are both 1 / 2 of the face collection module. When the first rotating shaft and the second rotating shaft respectively control the first voice collection module and the second voice collection module to rotate toward the first surface of the face collection module, the second surface and the third surface gradually approach the first surface, and finally cling to the first surface, so that the face collection module, the first voice collection module and the second voice collection module are nested into a rectangular structure, and the face recognition sensor, the first voice recognition sensor and the second voice recognition sensor are hidden in the space surrounded by the first surface, the second surface and the third surface, thereby improving the dustproof performance of the biometric information collection mechanism when not in use.

[0038] In an implementation of the first aspect, the movable support mechanism further includes a base;

[0039] The base is a hollow structure in the middle, and the telescopic rod is plugged into the base.

[0040] In the above-mentioned biometric identification and collection device, the base of the movable support mechanism is plugged into the telescopic rod to fix the movable support mechanism and improve the stability of the biometric identification and collection device.

[0041] In a second aspect, the present application provides a conference minutes device, including a display and the biometric identification and collection device described in the first aspect; wherein the display is connected to a data processor of the biometric identification and collection device.

[0042] In the above-mentioned meeting minutes device, the data processor of the biometric identification and collection device inputs the meeting minutes generated by the facial and voice data into the display through the second output interface, thereby realizing the visualization of the meeting minutes.

[0043] It can be seen from the above technical scheme that the present application provides a biometric identification and collection device and a meeting minutes device, the device comprising: a biometric information collection mechanism, a movable support mechanism and a data processor; the biometric information collection mechanism comprises a face collection module, a first voice collection module, a second voice collection module, a first rotating shaft and a second rotating shaft, the face collection module, the first voice collection module, the second voice collection module and the data processor are electrically connected; the face collection module, the first voice collection module and the second voice collection module are rectangular structures, the first rotating shaft and the second rotating shaft are cylindrical structures; the first rotating shaft is arranged on the face collection module On one side of the block, the first rotating shaft is rotatably connected to the face acquisition module and the first voice acquisition module, and the second rotating shaft is set on the other side of the face acquisition module, and the second rotating shaft is rotatably connected to the face acquisition module and the second voice acquisition module; the movable supporting mechanism includes a connecting rod, a rotating joint and a telescopic rod, and the movable supporting mechanism is electrically connected to the data processor; the connecting rod is a cylindrical structure, the rotating joint is a spherical structure, the telescopic rod includes a first sliding rod and a second sliding rod, the first sliding rod and the second sliding rod are cylindrical structures, and the first sliding rod is slidably connected to the second sliding rod; the connecting rod is connected to the biological information acquisition mechanism, and the rotating joint is rotatably connected to the connecting rod and the telescopic rod.

[0044] The face acquisition module in the biometric information acquisition mechanism determines the identity information of the speaker according to the face recognition result. The first voice acquisition module and the second voice acquisition module are respectively arranged on both sides of the face acquisition module to improve the accuracy of sound source positioning and voice recognition. The first rotating shaft and the second rotating shaft drive the first voice acquisition module and the second voice acquisition module to rotate, and control the expansion and closing states of the first voice acquisition module and the second voice acquisition module. The rotating joint in the movable support mechanism drives the connecting rod to rotate, so that the connecting rod drives the biometric information acquisition mechanism to adjust the angle, and the telescopic rod drives the biometric information acquisition mechanism to adjust the height. Through the angle and height adjustment, the biometric information acquisition mechanism can face the speaker's face, thereby improving the accuracy of recognizing the speaker's facial information and voice information. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solution of the present application, the drawings required for use in the embodiments are briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0046] Figure 1 This is a schematic diagram of the structure of the biometric identification and collection device described in an embodiment of the present application;

[0047] Figure 2 A schematic diagram of the positional relationship between a speaker and a biometric information collection mechanism according to an embodiment of the present application;

[0048] Figure 3 This is another schematic diagram of the positional relationship between a speaker and a biometric information collection mechanism according to an embodiment of the present application;

[0049] Figure 4 A schematic diagram of transmission between the biometric identification and collection device and an external device described in an embodiment of the present application;

[0050] Figure 5 This is a structural schematic diagram of the closed state of the biological information collection mechanism described in the embodiment of the present application;

[0051] Figure 6 This is a schematic diagram of the operation flow of the biometric identification and collection device described in an embodiment of the present application.

[0052] Illustration Description:

[0053] Among them: 1-biological information acquisition mechanism; 101-face acquisition module; 1011-first surface; 1012-face recognition sensor; 102-first voice acquisition module; 1021-second surface; 1022-first voice recognition sensor; 103-second voice acquisition module; 1031-third surface; 1032-second voice recognition sensor; 104-first rotating shaft; 105-second rotating shaft; 2-movable supporting mechanism; 201-connecting rod; 202-rotating joint; 203-telescopic rod; 2031-first sliding rod; 2032-second sliding rod; 301-speaker; 401-display. DETAILED DESCRIPTION

[0054] The following embodiments are described in detail, and examples thereof are shown in the accompanying drawings. When the following description refers to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The implementations described in the following embodiments do not represent all implementations consistent with the present application. They are only examples of systems and methods consistent with some aspects of the present application as detailed in the claims.

[0055] It should be noted that the brief description of terms in this application is only for the convenience of understanding the embodiments described below, and is not intended to limit the embodiments of this application. Unless otherwise specified, these terms should be understood according to their common and usual meanings.

[0056] The terms "first", "second", "third", etc. in the specification and claims of this application and the above drawings are used to distinguish similar or similar objects or entities, and do not necessarily mean to limit a specific order or sequence, unless otherwise noted. It should be understood that the terms used in this way can be interchangeable under appropriate circumstances.

[0057] In a meeting scenario, you can use a meeting minutes device to record and summarize the meeting content. A meeting minutes device is an intelligent device that integrates recording, video recording, and voice-to-text functions to help users record meeting content and provide support for subsequent review, analysis, and decision-making of meeting content.

[0058] The conference minutes device uses voice recognition technology to collect voice signals at the meeting site through voice recognition sensors, and performs voice recognition and transcription through algorithms. This device can accurately recognize and record voice content in scenarios where there is a single speaker or multiple speakers at a close distance.

[0059] However, when facing multiple speakers and at different distances from each speaker, the conference record device cannot accurately identify the voice information of each speaker. In addition, the conference record device is easily interfered by environmental factors when collecting voice information. For example, when the angle of the conference record device is not facing the face of the current speaker, the voice recognition sensor installed on the conference record device will not only collect the speaker's voice signal, but also collect interference information such as noise in the environment, reducing the accuracy of voice and voiceprint recognition.

[0060] In order to solve the problem that the angle of the conference minutes device is fixed and cannot be adjusted according to the position of the speaker, and the voice and voiceprint recognition accuracy is low, the embodiment of the present application provides a biometric recognition and collection device and a conference minutes device. Figure 1 This is a schematic diagram of the structure of the biometric identification and collection device described in the embodiment of the present application. Figure 2 This is a schematic diagram of the positional relationship between a speaker and a biometric information collection mechanism according to an embodiment of the present application. Figure 3 This is another schematic diagram of the positional relationship between a speaker and a biometric information collection mechanism according to an embodiment of the present application. Figure 4 This is a schematic diagram of the transmission between the biometric identification and collection device and the external device described in the embodiment of the present application. Figure 5 This is a schematic diagram of the structure of the closed state of the biological information collection mechanism described in the embodiment of the present application. Figure 6The following is a schematic diagram of the operation flow of the biometric identification and collection device described in the embodiment of the present application. Figure 1-Figure 6 The biometric identification and collection device and meeting minutes equipment provided in this application are described in detail.

[0061] The biometric identification and collection device provided in the embodiment of the present application includes: a biometric information collection mechanism 1, a movable support mechanism 2 and a data processor; the biometric information collection mechanism 1 includes a face collection module 101, a first voice collection module 102, a second voice collection module 103, a first rotating shaft 104 and a second rotating shaft 105, and the face collection module 101, the first voice collection module 102, and the second voice collection module 103 are electrically connected to the data processor; the face collection module 101, the first voice collection module 102, and the second voice collection module 103 are rectangular parallelepiped structures, and the first rotating shaft 104 and the second rotating shaft 105 are cylindrical structures; the first rotating shaft 104 is arranged on one side of the face collection module 101, and the first rotating shaft 104 is rotatably connected to the face collection module 101 and the first rotating shaft 104. A voice collection module 102, a second rotating shaft 105 is arranged on the other side of the face collection module 101, and the second rotating shaft 105 is rotatably connected to the face collection module 101 and the second voice collection module 103; the movable support mechanism 2 includes a connecting rod 201, a rotating joint 202 and a telescopic rod 203, and the movable support mechanism 2 is electrically connected to the data processor; the connecting rod 201 is a cylindrical structure, the rotating joint 202 is a spherical structure, and the telescopic rod 203 includes a first sliding rod 2031 and a second sliding rod 2032, the first sliding rod 2031 and the second sliding rod 2032 are cylindrical structures, and the first sliding rod 2031 is slidably connected to the second sliding rod 2032; the connecting rod 201 is connected to the biological information collection mechanism 1, and the rotating joint 202 is rotatably connected to the connecting rod 201 and the telescopic rod 203.

[0062] For example, see Figure 1 The first rotating shaft 104 drives the first voice acquisition module 102 to rotate, and the second rotating shaft 105 drives the second voice acquisition module 103 to rotate, so that the face acquisition module 101, the first voice acquisition module 102 and the second voice acquisition module 103 in the biological information acquisition mechanism 1 are oriented in the same direction. The connecting rod 201 in the movable supporting mechanism 2 is connected to the biological information acquisition mechanism 1. The rotating joint 202 rotates to connect the connecting rod 201 and the telescopic rod 203, which can drive the connecting rod 201 to rotate and thus adjust the angle of the biological information acquisition mechanism 1. The first sliding rod 2031 in the telescopic rod 203 is slidably connected to the second sliding rod 2032, which can adjust the height of the telescopic rod 203 and thus drive the biological information acquisition mechanism 1 to adjust its height.

[0063] In some embodiments, the face acquisition module 101 includes a first surface 1011 and a face recognition sensor 1012, the first surface 1011 is a rectangle, the face recognition sensor 1012 is a nested structure, and the face recognition sensor 1012 is arranged on the first surface 1011; the first voice acquisition module 102 includes a second surface 1021 and a first voice recognition sensor 1022, the second surface 1021 is a rectangle, the outer shell of the first voice recognition sensor 1022 is provided with a hollow structure, and the first voice recognition sensor 1022 is arranged on the second surface 1021; the second voice acquisition module 103 includes a third surface 1031 and a second voice recognition sensor 1032, the third surface 1031 is a rectangle, the outer shell of the second voice recognition sensor 1032 is provided with a hollow structure, and the second voice recognition sensor 1032 is arranged on the third surface 1031.

[0064] For example, see Figure 1 The face recognition sensor 1012 is disposed on the first surface 1011 of the face acquisition module 101, the first voice recognition sensor 1022 is disposed on the second surface 1021 of the first voice acquisition module 102, and the second voice recognition sensor 1032 is disposed on the third surface 1031 of the second voice acquisition module 103. The position of the first voice recognition sensor 1022 on the first voice acquisition module 102 and the position of the second voice recognition sensor 1032 on the second voice acquisition module 103 are symmetrical to improve the accuracy of sound source localization, voiceprint recognition and voice recognition.

[0065] It can be understood that the positions of the face recognition sensor 1012, the first voice recognition sensor 1022 and the second voice recognition sensor 1032 are not fixed, and can be respectively set at other positions on the face acquisition module 101, the first voice acquisition module 102 and the second voice acquisition module 103, as long as the first voice acquisition module 101 and the second voice acquisition module 102 are unobstructed when they are in a closed state, and the embodiments of the present application do not make specific limitations.

[0066] In some embodiments, the face recognition sensor 1012 includes a camera module; the data processor includes a first input interface; and the camera module is electrically connected to the first input interface.

[0067] The camera module collects facial information of the speaker 301 during the meeting, and inputs the facial information into the data processor via the first input interface. The data processor identifies the acquired facial information, matches it with the stored facial information, and obtains the identity information of the speaker 301.

[0068] Exemplarily, a camera module captures a face image and uses methods such as CNN (Convolutional Neural Networks) to locate the face in the image; the face image is then preprocessed to improve image quality and reduce noise; after preprocessing, key facial features such as facial contour, eyes, nose, and mouth are further extracted and the distance, angle, etc. between these facial features are measured based on geometric features; features are then extracted through template matching or model fitting and the extracted features are converted into an encoded form using methods such as principal component analysis; finally, the similarity between the extracted features and the features in the database is calculated using methods such as cosine similarity. If the similarity exceeds a first threshold, the match is successful and the corresponding identity information is output.

[0069] In some embodiments, the first speech recognition sensor 1022 includes a first microphone array; the data processor also includes a second input interface; and the first microphone array is electrically connected to the second input interface.

[0070] In some embodiments, the second speech recognition sensor 1032 includes a second microphone array; the data processor also includes a third input interface; and the second microphone array is electrically connected to the third input interface.

[0071] For example, see Figure 1 The first microphone array in the first voice recognition sensor 1022 is mainly used to collect voice information within the coverage of the first voice collection module 102, and the second microphone array in the second voice recognition sensor 1032 is mainly used to collect voice information within the coverage of the second voice collection module 103. The first microphone array and the second microphone array respectively input the collected voice signals into the data processor through the second input interface and the third input interface for processing, which can realize the all-round collection and processing of the sound source at the conference site, and improve the collection efficiency and accuracy of the voice signals.

[0072] In some embodiments, the movable support mechanism 2 also includes a first drive motor, a second drive motor, a reducer and a screw. The first output shaft of the first drive motor is connected to the rotation joint 202 through the reducer, and the second output shaft of the second drive motor is connected to the telescopic rod 203 through the screw; the data processor also includes a first output interface; the first drive motor, the second drive motor and the first output interface are electrically connected.

[0073] After the first voice acquisition module 102 and the second voice acquisition module 103 preliminarily acquire the voice information of the speaker 301, the voice information is input into the data processor for processing to obtain the location information of the speaker 301. For example, based on the sound source localization technology of TDOA (Time Difference Of Arrival), the time difference between the arrival of the signal at the first microphone array and the second microphone array is estimated, thereby obtaining the equation group of the sound source position coordinates. The azimuth coordinates of the sound source are determined by solving the equation group to realize the sound source localization. Then, according to the location information of the speaker 301, the adjustment signal for the rotary joint 202 and the telescopic rod 203 in the movable support mechanism 2 is obtained, and the adjustment signal is output to the first drive motor and the second drive motor of the movable support mechanism 2 through the first output interface, and the first output shaft of the first drive motor transmits the rotation to the rotary joint 202 through the reducer, driving the rotary joint 202 to rotate, and the second output shaft of the second drive motor converts the rotation into linear motion through the lead screw and transmits it to the telescopic rod 203, driving the telescopic rod 203 to telescope.

[0074] For example, see Figure 2 , the location of the speaker 301 and the current location of the biometric information collection agency 1 are as follows Figure 2 As shown, the face acquisition module 101, the first voice acquisition module 102 and the second voice acquisition module 103 in the biometric information acquisition mechanism 1 are not facing the face of the speaker 301. After receiving the voice signal, the data processor processes the voice signal to obtain the position information of the speaker 301, and then obtains the adjustment signal and outputs the adjustment signal to the first drive motor and the second drive motor in the movable support mechanism 2. After receiving the adjustment signal, the first output shaft of the first drive motor drives the rotary joint 202 to rotate through the reducer to drive the biometric information acquisition mechanism 1 to rotate counterclockwise to the angle where the face acquisition module 101 is facing the face of the speaker 301; then, the second output shaft of the second drive motor drives the first slide bar 2031 in the telescopic rod 203 to slide downward through the lead screw to drive the biometric information acquisition mechanism 1 to move downward to the same horizontal line as the face of the speaker 301.

[0075] The adjusted biometric identification and collection device is as follows Figure 3 As shown, the face acquisition module 101 in the biometric information acquisition mechanism 1 is facing the face of the speaker 301 , and the face acquisition module 101 , the first voice acquisition module 102 and the second voice acquisition module 103 are all on the same horizontal line with the face of the speaker 301 .

[0076] In some embodiments, the data processor further includes a second output interface; the second output interface is electrically connected to an external device.

[0077] For example, see Figure 4 The display 401 is an external device, which is electrically connected to the data processor through the second output interface. The biometric identification and collection device collects the face information and voice information of the speaker 301, and performs face recognition through CNN and other methods to obtain the identity information of the speaker 301, and performs voice recognition through HMM (Hidden Markov Model) and other methods to convert the voice content into text content, obtain the meeting minutes, and output them to the display 401 through the second output interface of the data processor for display.

[0078] It is understandable that the external device may also be other devices that can be used to store or display meeting minutes, such as a removable hard disk, etc., and the embodiments of the present application do not make specific limitations.

[0079] In some embodiments, the length of the second surface 1021 is 1 / 2 of the length of the first surface 1011 , and the height of the second surface 1021 is equal to the height of the first surface 1011 ; the length of the third surface 1031 is 1 / 2 of the length of the first surface 1011 , and the height of the third surface 1031 is equal to the height of the first surface 1011 .

[0080] For example, see Figure 1 The first rotating shaft 104 and the second rotating shaft 105 respectively control the first voice acquisition module 102 and the second voice acquisition module 103 to rotate toward the first surface 1011 in the face acquisition module 101, and the second surface 1021 and the third surface 1031 are finally close to the first surface 1011, so that the face acquisition module 101, the first voice acquisition module 102 and the second voice acquisition module 103 are nested into a rectangular structure. At this time, the biological information acquisition mechanism 1 is in a closed state.

[0081] The biological information collection mechanism 1 in the closed state is as follows Figure 5 As shown, the face recognition sensor 1012 , the first voice recognition sensor 1022 , and the second voice recognition sensor 1023 are hidden in a space surrounded by the first surface 1011 , the second surface 1021 , and the third surface 1031 .

[0082] In some embodiments, the movable support mechanism 2 further includes a base 204 ; the base 204 is a hollow structure in the middle, and the telescopic rod 203 is plugged into the base 204 .

[0083] It is understandable that the telescopic rod 203 and the base 204 can also be connected by other means, such as bonding, etc., as long as the telescopic rod and the base are fixed to improve the stability of the biometric identification and collection device, and the embodiments of the present application do not make specific limitations.

[0084] Some embodiments of the present application also provide a meeting minutes device, including the biometric identification and collection device and a display 401 described in the above embodiments; wherein the display 401 is connected to a data processor of the biometric identification and collection device.

[0085] The meeting minutes device provided by this application is described below with reference to a specific example. Figure 6 This is a schematic diagram of the operation flow of the biometric identification and collection device described in the embodiment of this application. Figure 6 The operation process includes the following steps:

[0086] S601, the voice collection module is expanded.

[0087] For example, see Figure 5 The biometric information collection mechanism 1 is in a closed state in an initial state. When the meeting starts, the first voice collection module 102 and the second voice collection module 103 rotate in a direction away from the first surface 1011 driven by the first rotating shaft 104 and the second rotating shaft 105 respectively.

[0088] S602: Expose a face acquisition module and a voice acquisition module.

[0089] For example, see Figure 1 When the first voice collection module 102 and the second voice collection module 103 are in the unfolded state, the face collection module 101 is exposed. At this time, the face recognition sensor 1012 in the face collection module 101, the first voice recognition sensor 1022 in the first voice collection module 102, and the second voice recognition sensor 1032 in the second voice collection module 103 are all exposed, and face information and voice information can be collected.

[0090] S603: localize the sound source.

[0091] When voice information is detected, the first microphone array in the first voice recognition sensor 1022 and the second microphone array in the second voice recognition sensor 1032 work together to estimate the time difference between the signal arriving at the first microphone array and the second microphone array based on the TDOA (Time Difference Of Arrival) sound source localization technology, thereby obtaining a set of equations for the coordinates of the sound source position. The azimuth coordinates of the sound source are determined by solving the set of equations to achieve sound source localization.

[0092] S604: Automatically adjust the height and angle of the module so that it faces the speaker.

[0093] For example, see Figure 2, the face acquisition module 101, the first voice acquisition module 102 and the second voice acquisition module 103 in the biometric information acquisition mechanism 1 are not facing the face of the speaker 301. After the sound source is located, the first drive motor and the second drive motor in the movable support mechanism 2 receive the adjustment signal sent by the data processor, and the first output shaft of the first drive motor drives the rotary joint 202 to rotate through the reducer to drive the biometric information acquisition mechanism 1 to rotate counterclockwise to the angle that the face acquisition module 101 faces the face of the speaker 301; then, the second output shaft of the second drive motor drives the first slide bar 2031 in the telescopic rod 203 to slide downward through the lead screw to drive the biometric information acquisition mechanism 1 to move downward to the same horizontal line as the face of the speaker 301.

[0094] The adjusted biometric identification and collection device is as follows Figure 3 As shown, the face acquisition module 101 in the biometric information acquisition mechanism 1 is facing the face of the speaker 301 , and the face acquisition module 101 , the first voice acquisition module 102 and the second voice acquisition module 103 are all on the same horizontal line with the face of the speaker 301 .

[0095] S60511. Collect and identify the speaker’s face and voiceprint.

[0096] For example, see Figure 3 When the biometric information collection mechanism 1 is facing the face of the speaker 301, the face collection module 101 collects the facial information of the speaker 301 and inputs it into the data processor through the first input interface, and processes it using methods such as CNN; the first voice collection module 102 and the second voice collection module 103 jointly collect the voice information of the speaker 301, and input it into the data processor through the second input interface and the third input interface, and processes it using methods such as HMM.

[0097] S60512. Determine the speaker's identity based on the recognition result.

[0098] After receiving the facial information, the data processor matches the facial information with the stored face information to obtain the facial identity information of the speaker 301; after receiving the voice information, the data processor extracts the voiceprint information of the speaker 301, matches the extracted voiceprint information with the stored voiceprint information to obtain the voiceprint identity information of the speaker 301; the facial identity information is combined with the voiceprint identity information to obtain the final identity information of the speaker 301.

[0099] S60521. Collect and recognize the speaker's voice.

[0100] For example, see Figure 3When the biometric information collection mechanism 1 is facing the face of the speaker 301, the first voice collection module 102 and the second voice collection module 103 jointly collect the voice information of the speaker 301, and input it into the data processor through the second input interface and the third input interface, and process it using methods such as HMM.

[0101] S60522. Transcribe the speech text according to the recognition result.

[0102] After receiving the voice information, the data processor first converts the sound signal into a digital signal and performs noise reduction processing; then extracts parameters that can represent the voice features from the processed voice signal, and maps the extracted voice feature parameters to possible words or phrases; then predicts the most likely word sequence based on grammatical rules and vocabulary knowledge; finally, the possible words or phrases and the most likely word sequence are combined to find the most likely text sequence. For example, the data processor can run HMM to identify text sequences from voice signals.

[0103] S606: Combine the speaker's identity and the voice text to form meeting minutes.

[0104] After obtaining the final identity information of the speaker 301 and converting the speech of the speaker 301 into text, the final identity information is combined with the text content to form the final meeting minutes.

[0105] For example, see Figure 4 ,The final generated meeting minutes can be displayed on the monitor in a visual form for ,convenient review and discussion.

[0106] It is understandable that during the meeting, each time a speaker is replaced, the operation process of the above-mentioned biometric identification and collection device can be repeated until the end of the meeting.

[0107] It can be seen from the above technical scheme that the present application provides a biometric identification and collection device and a meeting minutes device, the device comprising: a biometric information collection mechanism 1, a movable support mechanism 2 and a data processor; the biometric information collection mechanism 1 comprises a face collection module 101, a first voice collection module 102, a second voice collection module 103, a first rotating shaft 104 and a second rotating shaft 105, the face collection module 101, the first voice collection module 102, the second voice collection module 103 are electrically connected to the data processor; the face collection module 101, the first voice collection module 102 and the second voice collection module 103 are rectangular parallelepiped structures, the first rotating shaft 104 and the second rotating shaft 105 are cylindrical structures; the first rotating shaft 104 is arranged on one side of the face collection module 101, and the first rotating shaft 104 is rotated to connect the face collection module 101 to the first voice collection module 102. The face acquisition module 101 and the first voice acquisition module 102, the second rotating shaft 105 is arranged on the other side of the face acquisition module 101, and the second rotating shaft 105 is rotatably connected to the face acquisition module 101 and the second voice acquisition module 103; the movable support mechanism 2 includes a connecting rod 201, a rotating joint 202 and a telescopic rod 203, and the movable support mechanism 2 is electrically connected to the data processor; the connecting rod 201 is a cylindrical structure, the rotating joint 202 is a spherical structure, and the telescopic rod 203 includes a first sliding rod 2031 and a second sliding rod 2032, the first sliding rod 2031 and the second sliding rod 2032 are cylindrical structures, and the first sliding rod 2031 is slidably connected to the second sliding rod 2032; the connecting rod 201 is connected to the biological information acquisition mechanism 1, and the rotating joint 202 is rotatably connected to the connecting rod 201 and the telescopic rod 203.

[0108] The face acquisition module in the biometric information acquisition mechanism determines the identity information of the speaker according to the face recognition result. The first voice acquisition module and the second voice acquisition module are respectively arranged on both sides of the face acquisition module to improve the accuracy of sound source positioning and voice recognition. The first rotating shaft and the second rotating shaft drive the first voice acquisition module and the second voice acquisition module to rotate, and control the expansion and closing states of the first voice acquisition module and the second voice acquisition module. The rotating joint in the movable support mechanism drives the connecting rod to rotate, so that the connecting rod drives the biometric information acquisition mechanism to adjust the angle, and the telescopic rod drives the biometric information acquisition mechanism to adjust the height. Through the angle and height adjustment, the biometric information acquisition mechanism can face the speaker's face, thereby improving the accuracy of recognizing the speaker's facial information and voice information.

[0109] Similar parts between the embodiments provided in this application can be referenced to each other. The specific implementation methods provided above are only a few examples under the general concept of this application and do not constitute a limitation on the protection scope of this application. For those skilled in the art, any other implementation methods expanded based on the scheme of this application without creative work belong to the protection scope of this application.

Claims

1. A biometric identification and collection device, characterized in that: include: A biological information collection mechanism (1), an activity support mechanism (2) and a data processor; The biometric information collection mechanism (1) comprises a face collection module (101), a first voice collection module (102), a second voice collection module (103), a first rotating shaft (104) and a second rotating shaft (105); the face collection module (101), the first voice collection module (102) and the second voice collection module (103) are electrically connected to the data processor; The face acquisition module (101), the first voice acquisition module (102) and the second voice acquisition module (103) are rectangular parallelepiped structures, and the first rotating shaft (104) and the second rotating shaft (105) are cylindrical structures; The first rotating shaft (104) is arranged at one side of the face acquisition module (101), and the first rotating shaft (104) is rotatably connected to the face acquisition module (101) and the first voice acquisition module (102); the second rotating shaft (105) is arranged at the other side of the face acquisition module (101), and the second rotating shaft (105) is rotatably connected to the face acquisition module (101) and the second voice acquisition module (103); The movable support mechanism (2) comprises a connecting rod (201), a rotating joint (202) and a telescopic rod (203), and the movable support mechanism (2) is electrically connected to the data processor; The connecting rod (201) is a cylindrical structure, the rotating joint (202) is a spherical structure, the telescopic rod (203) comprises a first sliding rod (2031) and a second sliding rod (2032), the first sliding rod (2031) and the second sliding rod (2032) are cylindrical structures, and the first sliding rod (2031) is slidably connected to the second sliding rod (2032); The connecting rod (201) is connected to the biological information collection mechanism (1), and the rotating joint (202) rotatably connects the connecting rod (201) and the telescopic rod (203).

2. The biometric identification and collection device according to claim 1, characterized in that: The face acquisition module (101) comprises a first surface (1011) and a face recognition sensor (1012), the first surface (1011) is a rectangle, the face recognition sensor (1012) is a nested structure, and the face recognition sensor (1012) is arranged on the first surface (1011); The first voice collection module (102) comprises a second surface (1021) and a first voice recognition sensor (1022); the second surface (1021) is rectangular; a hollow structure is provided on the outer shell of the first voice recognition sensor (1022); and the first voice recognition sensor (1022) is arranged on the second surface (1021); The second voice collection module (103) comprises a third surface (1031) and a second voice recognition sensor (1032); the third surface (1031) is rectangular; a hollow structure is provided on the outer shell of the second voice recognition sensor (1032); and the second voice recognition sensor (1032) is arranged on the third surface (1031).

3. The biometric identification and collection device according to claim 2, characterized in that: The face recognition sensor (1012) comprises a camera module; The data processor comprises a first input interface; The camera module is electrically connected to the first input interface.

4. The biometric identification and collection device according to claim 2, characterized in that: The first speech recognition sensor (1022) includes a first microphone array; The data processor also includes a second input interface; The first microphone array is electrically connected to the second input interface.

5. The biometric identification and collection device according to claim 2, characterized in that: The second speech recognition sensor (1032) includes a second microphone array; The data processor further comprises a third input interface; The second microphone array is electrically connected to the third input interface.

6. The biometric identification and collection device according to claim 1, characterized in that: The movable support mechanism (2) further comprises a first drive motor, a second drive motor, a reducer and a lead screw, wherein a first output shaft of the first drive motor is connected to the rotary joint (202) via the reducer, and a second output shaft of the second drive motor is connected to the telescopic rod (203) via the lead screw; The data processor also includes a first output interface; The first driving motor and the second driving motor are electrically connected to the first output interface.

7. The biometric identification and collection device according to claim 1, characterized in that: The data processor also includes a second output interface; The second output interface is electrically connected to an external device.

8. The biometric identification and collection device according to claim 2, characterized in that: The length of the second surface (1021) is 1 / 2 of the length of the first surface (1011), and the height of the second surface (1021) is equal to the height of the first surface (1011); The length of the third surface (1031) is 1 / 2 of the length of the first surface (1011), and the height of the third surface (1031) is equal to the height of the first surface (1011).

9. The biometric identification and collection device according to claim 1, characterized in that: The movable support mechanism (2) also includes a base (204); The base (204) is a hollow structure in the middle, and the telescopic rod (203) is plugged into the base (204).

10. A conference minutes device, characterized in that: It comprises a display and the biometric identification and collection device according to any one of claims 1 to 9; the display is connected to a data processor of the biometric identification and collection device.

Images