Split type desktop conference terminal all-in-one
Technical Field
The invention belongs to the technical field of intelligent software and hardware equipment, and particularly relates to a split desktop conference terminal all-in-one machine.
Background
Desktop conference terminals in existing conference rooms generally have several forms:
Desktop omnidirectional microphones, such devices typically integrate multiple microphones to pick up human voice in a conference room, and generally serve as peripherals without an operating system, so that conference software cannot be run;
The conference terminal host can generally run conference software based on windows, android or other operating systems, but does not have audio and video capabilities, and needs to be accessed to peripherals to realize audio and video acquisition;
The conference large screen, the equipment integrates a screen, a camera and an array microphone, can meet the audio and video requirements of conference room scenes, is generally provided with an operating system, and can directly run conference software;
the multimedia conference terminal generally integrates a camera and an array microphone, is placed above a desktop or a television, integrates an operating system, can supply video signals to display equipment such as a television projector through HDMI, has audio and video and conference software operation capability, and is only used as an audio and video peripheral, and the display equipment or the terminal which needs to be accessed is provided with the operating system.
The prior art has the defects that:
The desktop microphone generally cannot directly run conference software, and needs to purchase a terminal or a large screen with a system for use, so that the complexity of system connection is increased and the use cost is increased;
the conference terminal host also needs to purchase peripheral equipment to realize the audio and video acquisition function, so that the connection complexity and cost are increased, a user cannot interact intuitively with the terminal, and interaction can only be performed through a mouse and a keyboard or a remote controller, so that the wiring of conference room equipment is disordered, and accessories are easy to lose;
The conference large screen is characterized in that the large screen is placed at a position which is generally far away from a meeting person, so that the effective pickup distance and the pickup effect of the array microphone are not as good as those of the omni-directional microphone placed on a desktop, and a user needs to start up to click the screen in operations such as operation silencing and volume adjustment, and the like, unlike the desktop omni-directional microphone which is directly operated on the desktop, the touch frame or the touch screen adopted by the conference large screen for realizing interaction improves the use cost of the user;
the multimedia conference terminal has the same problems as a conference large screen in audio acquisition, and has the problems of poor interaction experience and easy loss of accessories because of the fact that an external control panel is needed or a remote controller is used for controlling the multimedia conference terminal in interaction because of the fact that the conference large screen is not available.
Based on the above, the invention designs a split desktop conference terminal all-in-one machine to solve the above problems.
Disclosure of Invention
The invention aims to solve the problems of poor interaction experience, poor pickup effect, complex equipment wiring and high cost of easily lost accessories of the existing product, and provides a split desktop conference terminal all-in-one machine.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
The utility model provides a split type desktop meeting terminal all-in-one, includes the base, the top of base is provided with meeting host computer base, link to each other through rotary mechanism between base and the meeting host computer base, 8 microphones that are annular array are installed to the side-mounting of meeting host computer base for pick up the speaker's voice in the meeting, install the voice recognition system that is used for controlling 8 microphones to carry out voice recognition on the meeting host computer base, the touch panel support that charges is inhaled for taking magnetism to the top of meeting host computer base for place touch meeting accuse board and charge for it, touch meeting accuse board installs the self-research application software based on android system, realizes the control operation of meeting software through the touch.
As a further description of the above technical solution:
the conference host base freely rotates through the rotating mechanism, the rotating angle is not more than 360 degrees, the bottom of the base is provided with HDMI, USB, a network port and a power supply interface, the HDMI outputs audio and video signals of the PC host to conference display equipment, and the conference display equipment comprises a television, a projector and a conference large screen.
As a further description of the above technical solution:
the conference host base comprises a PC host with an X86 architecture, and the PC host is provided with a windows system for running conference software and self-research conference control software, and is provided with a high-fidelity loudspeaker system and an 8-array microphone system.
As a further description of the above technical solution:
the touch conference control panel is combined with the self-developed IOT device and used for realizing control operation of the IOT device related to the conference room, the touch conference control panel supports handwriting and is used for projecting handwriting contents on a touch screen to conference display equipment in the conference room, and the touch conference control panel is self-charged and used for being used separately from a conference host base.
As a further description of the above technical solution:
The touch can be controlled and install leading wide-angle camera on the flat board for videoconference, leading wide-angle camera can manual regulation pitch angle.
As a further description of the above technical solution:
the bottom of touch can accuse flat board is provided with the magnetism and inhales the contact that charges for charge to can accuse flat board through meeting host computer base, and have basic serial communication function, the meeting host computer base is last to correspond to go up the magnetism and inhale the contact that charges and be provided with down the magnetism and inhale the contact that charges, be used for charging to touch can accuse flat board through meeting host computer base, and have basic serial communication function.
As a further description of the above technical solution:
the number of the high-fidelity speakers is two, and the speakers are respectively distributed on the left side and the right side and are used for providing high-fidelity stereo sound effect.
As a further description of the above technical solution:
The voice recognition system comprises a ring microphone array, an FPGA, a Pansy voice processing module, a voice recognition module and a voice recognition output module;
The voice algorithm processing is carried out on the voice recordings of 8 paths of microphones by adopting a mode of 8-microphone annular layout to realize 2-6 m long-distance voice interaction, the microphones are provided with voice awakening functions, and a recording training mode is adopted to improve the awakening recognition rate and can be positioned to the speaker position so as to enable a conference host base to turn to the speaker, the annular microphone array forms pickup beams in the speaker direction, the speaker sound is enhanced, and surrounding background sounds and reverberation are suppressed;
The Pansy voice processing module adopts a specific hardware board card with annular layout.
As a further description of the above technical solution:
The voice recognition module is provided with an offline voice recognition engine, the offline voice recognition engine uses a smart cloud offline word list recognition technology, localized acoustic models and language models are processed on voices by loading the offline voice recognition engine and the offline voice package, the voice recognition module is connected with the offline voice recognition engine in a parallel mode, a keyword list is stored in the offline voice package of the recognition engine through the voice recognition module, the voice recognition process is also a process that the voice recognition module completes work, text content recognized through the voice recognition module is matched with keywords in the list, the keywords with the highest score are found out and are used as recognition results to be transmitted to voice recognition output processing, and the voice recognition output module plays corresponding prompt tones.
As a further description of the above technical solution:
the voice recognition model is also provided with a directional pickup problem model, the sound transmission process leads to periodic change of sound pressure, the microphone senses the sound pressure through the sensor and converts the sound pressure into an electric signal for sound pickup, the frequency domain signal model received by the microphone at t i is as follows, assuming that the sound source signal at t 0 is h (n):
Q(α,ti,t0)=P(ti,t0,α)V(α,t0)L(α)S(α)+Z(α);
Wherein,
The so-called stationary green's function, which is a sound field excited by a stationary point sound source at t 0, represents delay and attenuation caused by the distance between the sound source and the microphone, V (α, t 0) represents the directivity of the microphone, since the microphone is an omni-directional microphone whose frequency response is typically flat in the range of 20-2000Hz, V (α, t 0) is typically a constant, L (α) is the frequency response of the amplifier and ADC, and the frequency response of the high performance amplifier in the passband should be flat, and furthermore, the ADC typically employs noise shaping techniques such that the noise in the passband is small, thus in most cases it can be assumed that L (α) ≡1, z (α) is noise, typically including both correlated noise and uncorrelated noise.
In summary, due to the adoption of the technical scheme, the beneficial effects of the invention are as follows:
1. Compared with the split scheme of the traditional desktop microphone and the conference terminal, the device integrates the functions of a conference host and a control screen on the basis of the capability of the desktop microphone, and a user can have complete intelligent conference system experience without purchasing equipment.
2. Compared with the traditional conference large screen, the device has the advantages that the large screen is placed at a position which is generally far away from a conference person, so that the array microphone is provided with the array microphone, the effective pickup distance and the pickup effect of the array microphone are not as good as those of the omni-directional microphone placed on a desktop, a user is in mute operation, the volume adjustment operation needs to get up to click the screen, unlike the desktop omni-directional microphone which is directly operated on the desktop, the touch frame or the touch screen adopted by the conference large screen for realizing interaction is not convenient, the use cost of the user is increased, the device is placed on the desktop, the problems that the pickup distance is far away and the operation is inconvenient can be solved, and the device is provided with the touch screen, so that the operation similar to the large screen can be realized on a seat.
3. Compared with the traditional multimedia conference terminal integrated bar, the invention has the same problems as a conference large screen in audio acquisition, and in interaction, as the conference large screen is not provided with a touch screen, an external control panel is needed or a remote controller is used for controlling, thus the problems of poor interaction experience, complex connection and easy loss of accessories are solved.
4. In the invention, pansy boards with annular layout are adopted as the core of voice front-end processing, and are combined with corresponding offline voice recognition engines and voice recognition modules to be applied to a voice action control system of a service robot, and through the voice recognition tests of unspecific different distances, different angles and echo cancellation under a noise environment, the result shows that the system has higher recognition rate on remote commands under the noise environment, can eliminate echoes, is suitable for the application environment of a service robot, and is also suitable for the application of far-field and voice recognition systems under other noisy environments.
5. According to the invention, the voice recognition model is loaded with the directional pickup problem model, the collected data volume is large, the work is stable and reliable, the transmission rate is very high, the voice recognition model is suitable for directional pickup of a multichannel microphone array, and the voice recognition model can provide assistance for application of the microphone array.
Drawings
Fig. 1 is a schematic diagram of the overall structure of a split desktop conference terminal all-in-one machine according to the present invention;
fig. 2 is a frame diagram of a voice recognition structure in a split desktop conference terminal integrated machine according to the present invention;
fig. 3 is a schematic frame diagram of voice recognition in a split desktop conference terminal all-in-one machine according to the present invention;
Fig. 4 is a diagram of a microphone receiving signal model in a split desktop conference terminal all-in-one machine according to the present invention.
Legend description:
1. The touch control panel comprises a touch panel support, a conference host base, an array microphone, a base, a front wide-angle camera, an upper magnetic attraction charging contact, a lower magnetic attraction charging contact and a high-fidelity loudspeaker, wherein the touch panel support comprises a touch panel body, a touch panel support, the conference host base, the array microphone, the base, the front wide-angle camera, the upper magnetic attraction charging contact, the lower magnetic attraction charging contact and the high-fidelity loudspeaker.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1-4, the invention provides a split desktop conference terminal all-in-one machine, which comprises a base 5, wherein a conference host base 3 is arranged above the base 5, the base 5 is connected with the conference host base 3 through a rotating mechanism, 8 microphones in an annular array are arranged on the side surface of the conference host base 3 and used for picking up the voice of a speaker in a conference, a voice recognition system used for controlling the 8 microphones to perform voice recognition is arranged on the conference host base 3, a touch panel bracket 2 with magnetic attraction charging is arranged at the top of the conference host base 3 and used for placing and charging a touch conference control panel 1, self-grinding application software based on an android system is arranged on the touch conference control panel 1, and the control operation of the conference software is realized through touch.
As a further description of the above technical solution:
The conference host base 3 freely rotates through the rotating mechanism, the rotating angle is not more than 360 degrees, the bottom of the base 5 is provided with HDMI, USB, a network port and a power supply interface, the HDMI outputs audio and video signals of the PC host to conference display equipment, and the conference display equipment comprises a television, a projector and a conference large screen.
As a further description of the above technical solution:
the conference host base 3 comprises a PC host with an X86 architecture, and the PC host carries a windows system for running conference software and self-research conference control software, which has a hi-fi speaker 9 system and an 8-array microphone 4 system.
As a further description of the above technical solution:
The touch conference control panel 1 is combined with self-developed IOT equipment, and is used for realizing control operation of the IOT equipment related to a conference room, the touch conference control panel 1 supports handwriting, is used for projecting handwriting content on a touch screen to conference display equipment in the conference room, and the touch conference control panel 1 is self-charged and is used for being used in a split mode with a conference host base 3.
As a further description of the above technical solution:
the touch-controllable panel 1 is provided with a front wide-angle camera 6 for video conferences, and the front wide-angle camera 6 can manually adjust pitching angles.
As a further description of the above technical solution:
The bottom of touch meeting accuse flat board 1 is provided with magnetism and inhales charging contact 7 for charge to meeting accuse flat board through meeting host computer base 3, and have basic serial communication function, meeting host computer base 3 is gone up to correspond magnetism and inhale charging contact 7 and is provided with magnetism down and inhale charging contact 8, is used for charging to touching meeting accuse flat board 1 through meeting host computer base 3, and have basic serial communication function.
As a further description of the above technical solution:
The number of the hi-fi speakers 9 is two, and the two speakers are respectively distributed on the left and right sides for providing the hi-fi stereo sound effect.
As a further description of the above technical solution:
The voice recognition system comprises a ring microphone array, an FPGA, a Pansy voice processing module, a voice recognition module and a voice recognition output module;
The annular microphone array adopts an 8-microphone annular layout mode, carries out voice algorithm processing on the sound recordings of 8-way microphones, realizes 2-6 m long-distance voice interaction, has a voice awakening function, adopts a recording training mode, is used for improving the awakening recognition rate, can be positioned to the speaker position, enables the conference host base 3 to turn to the speaker, forms pickup beams in the speaker direction, enhances the speaker sound, and inhibits surrounding background sounds and reverberation;
pansy the voice processing module adopts a specific hardware board card with annular layout.
As a further description of the above technical solution:
The voice recognition module is provided with an offline voice recognition engine, the offline voice recognition engine uses a smart cloud offline word list recognition technology, the voice is processed by loading the offline voice recognition engine and an offline voice packet, a localization acoustic model and a language model are processed on the voice, the voice recognition module is connected with the offline voice recognition engine in a parallel mode, a keyword list is firstly stored in the offline voice packet of the recognition engine through the voice recognition module, the voice recognition process is also a process that the voice recognition module completes work, text content recognized through the voice recognition module is matched with keywords in the list, the keywords with the highest separation are found out and are used as recognition results to be transmitted to voice recognition output processing, and the voice recognition output module plays corresponding prompt tones.
As a further description of the above technical solution:
The voice recognition model is also provided with a directional pickup problem model, the sound transmission process leads to periodic change of sound pressure, the microphone senses the sound pressure through the sensor and converts the sound pressure into an electric signal to pick up the sound, the sound source signal at t 0 is assumed to be h (n), and the frequency domain signal model received by the microphone at t i is as follows:
Q(α,ti,t0)=P(ti,t0,α)V(α,t0)L(α)S(α)+Z(α);
Wherein,
The so-called stationary green's function, which is a sound field excited by a stationary point sound source at t 0, represents delay and attenuation caused by the distance between the sound source and the microphone, V (α, t 0) represents the directivity of the microphone, since the microphone is an omni-directional microphone whose frequency response is typically flat in the range of 20-2000Hz, V (α, t 0) is typically a constant, L (α) is the frequency response of the amplifier and ADC, and the frequency response of the high performance amplifier in the passband should be flat, and furthermore, the ADC typically employs noise shaping techniques such that the noise in the passband is small, thus in most cases it can be assumed that L (α) ≡1, z (α) is noise, typically including both correlated noise and uncorrelated noise.
The present invention is not limited to the above-mentioned embodiments, and any person skilled in the art, based on the technical solution of the present invention and the inventive concept thereof, can be replaced or changed within the scope of the present invention.