689 results about "Pickup" patented technology

The invention relates to a simulation digital human real-time intelligent voice interaction system and a simulation digital human real-time intelligent voice interaction method based on vision and a large model, and aims to solve the problems of inaccurate target speaker recognition, high response delay and the like in digital human voice interaction in a complex scene. The system circles an effective recognition range through a camera, triggers audio collection in combination with face detection, locks a target speaker and reduces noise by using lip movement recognition and sound image fusion technologies, converts the target speaker into a text through voice wake-up, generates an answer by means of a large language model (LLM) and knowledge retrieval enhancement (RAG) technologies, generates low-delay voice through a voice synthesis technology accelerated by the vLLM, and performs voice recognition on the target speaker. And driving the preloaded digital human image to synthesize a video stream and pushing the video stream to a front end for rendering in real time. Accurate pickup, low-delay interaction and rapid digital human image switching in a complex environment are realized, the accuracy and real-time performance of intelligent voice question answering are improved, and the method is suitable for government affair halls, exhibition halls and other scenes.

The invention provides a box body and a signal transmission method. The box body comprises an upper shell, a lower shell, a first microphone, keys, a rotating shaft and an antenna assembly, the lower shell and the upper shell are installed in a matched mode to form a containing cavity, the containing cavity comprises a first surface, opposite to the upper shell, in the lower shell, the lower shell comprises an annular side face connected with the first surface, and the annular side face comprises a first side face; the second side face and the third side face are connected with the first side face, the second side face and the third side face are oppositely arranged, a first microphone pickup hole is formed in the first side face, the first microphone is arranged in the lower shell, communicated with the first microphone pickup hole and used for collecting a first audio signal outside the box body, and the key is arranged on the second side face and used for collecting a second audio signal outside the box body. The rotating shaft is arranged corresponding to the third side surface and is connected between the lower shell and the upper shell, the upper shell and the lower shell are rotatably connected through the rotating shaft, and the antenna assembly is arranged on the first surface and is used for at least transmitting at least part of the first audio signal to the mobile terminal.

This disclosure relates to methods, systems, and techniques for visualizing external audio captured by a microphone of the vehicle. Using the techniques described herein, external audio signals may be interpreted and converted into clear visual indicium that can be understood by individuals associated with the vehicle, such as passengers inside the vehicle or individuals awaiting pickup.

The invention discloses a directional pickup method for a microphone. The microphone comprises a left microphone core, a middle microphone core and a right microphone core. Wherein the middle microphone core is located on the central axis of the microphone, and the left microphone core and the right microphone core are symmetrically distributed on the left side and the right side of the middle microphone core; sound of a sound source is synchronously collected through a left microphone core, a middle microphone core and a right microphone core of the microphone, and three paths of sound signals are obtained; respectively filtering the three paths of sound signals to obtain left, middle and right microphone core signals; calculating an included angle between a connecting line of a sound source and the middle microphone core and the central axis of the microphone based on the time difference and the distance difference of the sound reaching the left, middle and right microphone cores to obtain a sound source direction; based on the left microphone core signal, the middle microphone core signal, the right microphone core signal and the sound source direction, through time delay compensation and gain adjustment, the sound signal in the target direction is enhanced, the sound signal in the non-target direction is suppressed, the final output signal is obtained, and the directionality and the sound quality of pickup are improved.

The invention provides a power plant equipment sound abnormity identification and health prediction method based on granular computing and an LSTM network, and the method comprises the steps: employing an array pickup and a high-dynamic-range microphone array in a complex and high-noise background environment of a power plant, and combining an anti-interference filtering and beam forming algorithm, thereby achieving the sound abnormity identification and health prediction of the power plant equipment. According to the method, directional, multi-channel and non-contact sound acquisition is carried out on key parts of equipment, acquired sound signals are preprocessed, converted into time domain, frequency domain and time-frequency domain representations and coded into multi-dimensional numerical vectors, and compared with a rule-based expert system, the method has higher generalization ability and real-time response ability; a large language model is introduced, so that the output is closer to an engineering language and is suitable for operation and maintenance personnel to understand and execute; a self-defined knowledge base or safety semantic filtering is supported, and closed-loop deployment in an industrial field is facilitated; the system can be in butt joint with an intelligent inspection system, and full-link linkage of voice recognition, health assessment and suggestion generation is achieved.

The voice directional enhancement method comprises the steps that a current frame and a staring direction of mixed voice are acquired, and the mixed voice is acquired by a microphone array; constructing a direction feature based on the current frame of the mixed voice and the gazing direction, and constructing an audio feature based on the current frame of the mixed voice; determining a current frame mask according to the direction features and the audio features; and determining a target enhanced voice of the current frame according to the current frame mask, the current frame of the mixed voice and the plurality of historical frames of the mixed voice. In the embodiment of the invention, the mask of the current frame is generated under the guidance of the gazing direction, so that the target voice can be accurately separated from the mixed voice. Besides, based on the current frame mask and the current frame of the mixed voice, the target voice enhancement is performed frame by frame from the overall characteristics of the mixed voice by referring to a plurality of historical frames, so that the accuracy of the target enhanced voice of the current frame can be effectively improved, and thus the accuracy of directional pickup is improved.

The invention relates to a speech recognition method for a tactical AR helmet, and the method comprises the steps: collecting a first audio signal through a near-field pickup disposed near the mouth and nose of a wearer, and collecting a second audio signal through an environment pickup disposed at the outer side of the helmet or away from the mouth and nose; synchronously acquiring a third signal sounding with the wearer, wherein the third signal is at least one of helmet inner side contact type micro vibration, bone conduction vibration pickup or helmet inner sound pressure change; calculating a near-field difference indicator based on the first audio signal and the second audio signal, the near-field difference indicator comprising at least one of an energy difference or an energy ratio; determining sound production starting and ending moments from the first audio signal and the third signal, and calculating a time sequence synchronization index; and when the near-field difference index satisfies a preset near-field condition and the time sequence synchronization index satisfies a preset synchronization condition, determining that the voice is sent by the helmet wearer and a corresponding voice segment is output for subsequent identification, otherwise, shielding the voice segment.

The utility model relates to the technical field of pickup equipment, in particular to an intelligent acoustic positioning imager. The sound source positioning device comprises a lower shell, a sealing rubber mat and a microphone main board, the lower shell, the sealing rubber mat and the microphone main board are sequentially connected from outside to inside to form a sound source positioning assembly, and a plurality of digital silicon microphone structures are distributed on the sound source positioning assembly in an array mode. Each digital silicon microphone structure comprises a lower shell pickup port arranged on the lower shell, a rubber mat opening arranged on the sealing rubber mat and a mainboard pickup port arranged on the microphone mainboard, the microphone, the lower shell pickup port, the rubber mat opening and the mainboard pickup port are sequentially butted and communicated, the microphone is positioned on the microphone mainboard and is aligned with the mainboard pickup port, and the lower shell pickup port is positioned on the lower shell and is aligned with the mainboard pickup port. The digital silicon microphone structures are mutually independent. Multiple paths of high-sensitivity digital silicon microphones are arranged in an array mode to form a recording system for cooperative work, and the functions of sound source positioning, noise suppression and the like are achieved. And a stepped pickup channel is adopted, so that the direction and the distance to a sound source can be accurately positioned.

The invention discloses an intelligent pickup and speech recognition system based on multi-modal fusion, and relates to the technical field of artificial intelligence and speech recognition crossing. The system comprises a main control module, a plurality of syllable points and a multi-modal fusion engine, wherein the multi-modal fusion engine comprises four core components, namely a sound source localization and separation component, an environment adaptive noise reduction component, a cross-modal feature fusion component and a dynamic context understanding component. Multi-modal data are acquired through an array microphone and an auxiliary sensor group, and the system realizes sound source localization and separation, dynamic environment noise suppression, multi-modal feature deep fusion and context semantic correction. According to the method, the robustness, the accuracy and the intelligent interaction capability of voice recognition are effectively improved, the voice interaction experience is improved in complex scenes such as a noise environment and accent change, and a more reliable voice processing solution is provided for intelligent voice interaction equipment.

The invention relates to the technical field of directional pickup, and discloses a directional pickup method and device, computer equipment and a medium, and the method comprises the steps: obtaining time domain voice signals collected by at least two microphones, and carrying out the time-frequency conversion; calculating an observation phase difference of each frequency point, and determining a target theoretical phase difference according to a target pickup direction specified by a user and the geometric parameters of the microphone; encoding the target theoretical phase difference into a query vector, encoding the observation phase difference into a key vector, and generating a space confidence mask through a geometric cross attention mechanism; and filtering the frequency domain signal by using the mask, and outputting enhanced voice in a target direction through inverse time-frequency transformation. According to the invention, through an attention mechanism guided by physical prior, the problem of phase winding is effectively solved, and high-robustness and high-fidelity speech pickup in any direction is realized in a strong reverberation and low signal-to-noise ratio environment.

The invention relates to the technical field of microphone pickup, and discloses an intelligent pickup method and system of a microphone, and the method comprises the steps: respectively picking up an audio signal through a plurality of microphone units in a microphone array, carrying out the analysis of the audio signal to obtain a human voice signal and a background signal, and carrying out the collection of a voice signal based on the human voice signal and the background signal picked by each microphone unit. According to the technical scheme, the orientation angle of each microphone unit is adjusted, so that the microphone array is adjusted to the optimal pickup state, key frequency band recognition is performed on the human voice signal collected in the optimal pickup state, and voice enhancement processing is performed on the human voice signal according to the recognition result, so that the problem of poor pickup tracking capability of a microphone in the prior art is solved.

A microphone includes a shell assembly, a sound pickup assembly, a control assembly and a light-emitting assembly. The shell assembly includes a shell body with a first accommodating chamber. The sound pickup assembly includes a microphone head for sound pickup, wherein the microphone head is arranged on the shell assembly. The control assembly includes a first circuit board. The first circuit board is arranged in the first accommodating chamber, and is electrically connected to the microphone head. The light-emitting assembly includes a light-emitting unit. The light-emitting unit is arranged in the first accommodating chamber and is electrically connected to the first circuit board. The light-emitting unit is configured to allow emitted light to pass through a portion of the shell body.

This application provides a two-dimensional vector microphone, including: a wafer, a first pickup structure, and a second pickup structure; the first pickup structure and the second pickup structure are respectively disposed on opposite sides of the wafer in the thickness direction, or the first pickup structure and the second pickup structure are disposed on the same side of the wafer in the thickness direction. The first pickup structure extends along a first direction, and the second pickup structure extends along a second direction. The first direction and the second direction are orthogonal. The first pickup structure includes at least two parallel thermal resistance lines, and the second pickup structure includes at least two parallel thermal resistance lines. This application provides a two-dimensional vector microphone with good two-dimensional sound pickup effect.

The invention discloses a digital work card-driven business window multi-source recording data fusion intelligent analysis system, and particularly relates to the technical field of voice processing. The system is based on asynchronous audio streams collected by a wearable digital work card, an array microphone and environment pickup equipment; structured alignment and unified time reference construction of multi-source recording data are realized by adopting the steps of anchor point detection, event graph construction, cross-source matching, elastic time distortion alignment and the like. The system completes cross-source event marking by introducing multiple types of anchor point detectors (semantic keywords, acoustic abrupt changes and prompt tones); constructing an event alignment graph and executing confidence classification in combination with a unified clipping and cross sliding matching strategy in the anchor point pairing process; and then performing multi-scale alignment and resampling on the audio stream based on a layered elastic time warping method, and finally outputting a frame-level synchronous voice data stream to provide unified time support for downstream speaker separation, behavior auditing and semantic mapping.

The present application relates to the technical field of electronic devices. Disclosed is an electronic device, comprising a first shell, a second shell, a first microphone assembly, a display screen and a first sound pickup channel. The display screen and the first microphone assembly are arranged in the first shell, and the second shell surrounds and is connected to the first shell, and is partially located on the side of the display screen facing away from the first shell. A first inlet and a second inlet of the first sound pickup channel are in communication with the outside of the electronic device; the first inlet is provided in a first side portion of the electronic device and is enclosed by the first shell; the second inlet is enclosed by the first shell and / or the second shell; and an outlet of the first sound pickup channel is in communication with the first microphone assembly. In a first direction, compared with the orthographic projection of the first inlet on a first plane, the orthographic projection of the second inlet on the first plane is closer to the first microphone assembly, wherein the first plane is the plane where the direction of length and the direction of width of the electronic device are located, and the first direction is the direction from the first side portion to another side portion opposite the first side portion. Thus, the risk of hole blockage can be reduced, and the sound pickup performance is good.

The invention relates to the technical field of earphones, in particular to a microphone structure, an earphone and a test method. The microphone structure comprises a shell and a microphone assembly, the shell comprises a pickup hole, an exhaust hole and a containing cavity, the microphone assembly is located in the containing cavity, the microphone assembly and the outer wall of the shell are arranged in a spaced mode, the microphone assembly divides the containing cavity into a first cavity and a second cavity which are independent from each other, the pickup hole is communicated with the first cavity, and the exhaust hole is communicated with the second cavity. The earphone comprises the microphone structure. The testing method is used for testing the acoustic performance of the earphone. According to the microphone structure, the earphone and the test method, the noise reduction effect can be improved, and the definition of voice communication is ensured.

The pickup device (1) includes: an adaptive filter (141) that generates a speculative noise signal from a reference signal representing the noise signal component contained in the input signal acquired by the microphone (11); a noise removal signal generation unit (15) that generates a noise removal signal after subtracting the speculative noise signal from the input signal; a filter coefficient update unit (142) that updates the filter coefficients of the adaptive filter (141) using the noise removal signal; and a sample position determination unit (162) that determines at least one signal sample position from the signal sample position with the largest absolute value of the noise removal signal up to the predetermined largest signal sample position, wherein the filter coefficient update unit (142) updates the filter coefficients at the at least one signal sample position determined by the sample position determination unit (162).

The utility model provides magnetic type recording equipment. The magnetic type recording equipment comprises a shell; the sound receiving assembly is arranged in the shell, the sound receiving assembly comprises a sound guide part and a sound receiving part which are oppositely arranged, the sound guide part is provided with a sound guide hole communicated with the outside, the sound receiving part is provided with a sound receiving hole, the sound guide hole is communicated with the sound receiving hole to form a pickup cavity, and the diameter of the sound receiving hole is smaller than that of the sound guide hole; and the circuit board is arranged in the shell, the circuit board is provided with a connecting part, the connecting part is clamped between the sound guide part and the sound receiving part, and the circuit board is electrically connected with the sound receiving part. The structure of the sound recording equipment is simplified, the cost is reduced, the amount of environmental noise entering the sound guide piece is effectively reduced through the diameter difference between the sound receiving hole and the sound guide hole, and the sound recording definition and identification degree are improved.

The utility model discloses an earphone and playing equipment, and belongs to the technical field of earphones. The earphone comprises an earphone body and a shell, the earphone body is used for playing audio for the conchae cavity, the shell is connected with the earphone body, the shell is provided with a pickup hole, the pickup hole is used for picking up sound, the shell is provided with a pressure reduction area, the pressure reduction area is used for dispersing wind power of airflow, the pickup hole is formed in the pressure reduction area, and the opening direction of the pickup hole and the axial direction of the shell are arranged in a crossed mode. The earphone can reduce wind noise.

An information processing apparatus acquires a video captured by an imaging apparatus, acquires a plurality of sounds picked up by a plurality of sound pickup apparatuses in sync with capturing of the video, acquires state information regarding an attention state of a viewer to the video, and generates audio to be reproduced with the video by combining the plurality of sounds, wherein, in a case where the state information indicates that the viewer is having an overall view of the video, a plurality of sounds picked up by a plurality of sound pickup apparatuses set inside an area being watched by the viewer in the video are combined with an equal ratio, and sounds picked up by sound pickup apparatuses set outside the area are combined with a ratio that is reduced as the distance from the area to the sound pickup apparatuses increases.

The invention provides a data processing method and computer equipment, which can be applied to the field of audio and video processing, and comprises the following steps: determining first azimuth information of an abnormal event occurring in a physical space through a shooting device, and guiding the shooting device and a pickup device to perform linkage parameter adjustment according to the first azimuth information so as to focus a target area when the abnormal event occurs, the purpose of focusing is to only collect audios and videos in a target area. And when the sound source sounds in the target area, determining a first sub-area (the area where the person in the sound production state in the abnormal event is located) from the target area, and adjusting the pickup parameter again to enable the pickup range to be limited in the first sub-area. According to the method, firstly, the sound curtain of the area where the abnormal event is located is constructed through the parameter coarse adjustment process (namely, only audio and video in the target area are collected, and sound outside the target area is shielded), when the human voice is recognized in the target area, the parameters are further finely adjusted to achieve accurate directional pickup of the human voice in the abnormal event, and the audio and video evidence obtaining quality is effectively improved.

The utility model relates to the technical field of microphones, and discloses a portable sound pickup device capable of switching directivity, which comprises a gooseneck pipe and a sound pickup module, one end of the gooseneck pipe is connected with the sound pickup module, the other end of the gooseneck pipe is provided with a connecting port used for being plugged with electronic equipment, and the other end of the gooseneck pipe is connected with the sound pickup module. The pickup module comprises a rotary switching circuit unit, a mounting unit and a multi-direction pickup unit, the rotary switching circuit unit is arranged between the gooseneck pipe and the mounting unit, the multi-direction pickup unit is arranged on the mounting unit, the multi-direction pickup unit is electrically connected with the rotary switching circuit unit, and the rotary switching circuit unit is electrically connected with the mounting unit. The rotary switching circuit unit is electrically connected with the connecting port; the utility model has the following technical effects: the multi-angle pickup effect, multiple pickup modes and bending adjustment at different angles are realized.

The utility model discloses a kind of explanation equipment with electronic chest card function, explanation equipment includes shell, control mainboard, battery, sound pickup, display screen and clamping piece, shell has accommodating cavity;Control mainboard is set in accommodating cavity;Battery is set in accommodating cavity, and is electrically connected with control mainboard;Sound pickup is used to collect the voice of explanation personnel, is set in shell, and is electrically connected with control mainboard;Display screen is set in the front of shell, and is electrically connected with control mainboard, display screen can show the first display picture with explanation personnel chest card information, so that explanation equipment can be used as electronic chest card;Clamping piece is set in the back of shell, so that user can clamping wear explanation equipment.Explanation equipment can be used as electronic chest card while ensuring having interpreter function, both can satisfy work demand, and can keep appearance neat and unified, solve the inconvenience and unattractive problem caused by interpreter and chest card wearing simultaneously.

The application relates to the field of active noise reduction technology, in particular to an ANC (active noise control) active noise reduction simulation system, which comprises a first pickup sensor, a second pickup sensor, a simulation module, a microprocessor and a mediation unit. External sound of an ANC earphone is collected by the first pickup sensor, then noise reduction processing is carried out by using a noise reduction controller, internal sound of the ANC earphone is collected by the second pickup sensor, and waveform of the internal sound of the ANC earphone is simulated by the simulation module, then the waveform is compared with a preset optimal waveform by the microprocessor, parameters of a noise reduction filter are adjusted by the mediation unit, and features of the simulated sound waveform are extracted conveniently, the noise reduction module of the noise reduction earphone is systematically and visually adjusted, and noise reduction performance difference between left and right earphones is minimized.

Embodiments of the present application provide a kind of processing method and device of audio signal, it is related to media technical field, the method comprises: using N beams respectively to the audio signal to be handled is carried out beam processing, obtains N beam signals, N is greater than or equal to 2 integer;The audio signal to be handled is in target sound pickup range, and the audio signal including at least one sound source is collected by microphone, the coverage range of the N beams is in target sound pickup range;And determine N beam signals main beam signal and N-1 interference beam signals, the main beam signal is the beam signal of target sound source;And the adaptive filtering is carried out to main beam signal and N-1 interference beam signals, to obtain target sound source enhancement signal.Using the method provided in the embodiments of the present application, the interference signal in audio signal can be effectively suppressed, and the quality of audio signal is improved.

The invention discloses a conference privacy protection system and method based on visual sound field fusion. The conference privacy protection system comprises a visual acquisition module, an audio acquisition module, an audio playing module and a processing control module, wherein the visual acquisition module transmits mouth shape area three-dimensional coordinates, ear area three-dimensional coordinates and mouth motion states of participants to the processing control module; the processing control module generates a directional pickup instruction based on the mouth shape area three-dimensional coordinates, and sends the directional pickup instruction to the audio acquisition module to control the audio acquisition module to focus mouth shape area pickup; and meanwhile, a directional playback instruction is generated based on the three-dimensional coordinates of the ear region and is sent to the audio playing module to control the audio playing module to project sound beams to ears. Through visual guidance directional pickup, ultrasonic directional playback and audio and video fusion optimization, the risk of audio diffusion is reduced, so that the conference privacy protection effect is improved.

The invention discloses an open type wearable noise reduction earphone, and relates to the field of earphones. The earphone shell is provided with a pickup hole; the pickup hole is positioned on the inner side of the earphone shell; the microphone is arranged in the earphone shell and corresponds to the pickup hole; the noise reduction net-shaped structure is connected with the earphone shell and shields the pickup hole; and the noise reduction net structure is positioned on one side, close to the pickup hole, of the microphone. According to the microphone, the pickup hole is shielded by the noise reduction net structure, voice sound waves are allowed to pass through, the sensitivity of the microphone to effective voice signals is not remarkably reduced, and tone quality degradation or algorithm compensation requirement increase caused by blockage of a pickup channel is avoided; and the direct impact of wind noise on the microphone is avoided, physical weakening and energy dissipation are carried out on the airflow entering the pickup channel, and the influence of the wind noise on the pickup of the microphone is reduced from the structural level, so that the call quality in a wind noise scene is effectively improved on the premise of not influencing the open wearing experience.

The utility model discloses a high-quality noise reduction earphone, which comprises a shell, one end of the shell is provided with a sound outlet tube, and the end part of the sound outlet tube is provided with a sound outlet hole; the inner wall of the sound outlet tube is provided with a loudspeaker assembly and a feed-forward pickup assembly. The housing is also provided with a pickup channel structure and a control circuit board. The pickup channel structure comprises a first channel and a first pipeline. The pickup channel structure formed by the first channel and the first pipeline is used for directly collecting sound in the ear canal, the influence of sound filtering damping is prevented, and the first pipeline is arranged in the linear acoustic path design of the first pipeline in a linear acoustic path mode. The sound wave attenuation and the phase deviation are reduced, the influence of sound filtering damping is prevented, the pickup precision of the feed-forward pickup assembly on the sound in the ear canal is ensured, the subsequent noise reduction processing effect is improved, and the sound quality of the high-sound-quality noise reduction earphone is improved.

The invention discloses a distributed voice interaction method and electronic equipment in an outdoor scene, and the method comprises the steps: starting a distributed voice interaction terminal which comprises a pickup system with a microphone array, a positioning system and a closed-loop sound source tracking system; the sound source is preliminarily positioned through the pickup system; inputting the preliminary positioning result into a positioning system for accurate positioning; the positioning result is input into a closed-loop sound source tracking system, and the closed-loop sound source tracking system adjusts the pickup direction of the microphone in real time according to the positioning result, so that the microphone points to the sound source; and performing voice recognition based on the adjusted pickup signal to obtain voice interaction content of the sound source. The voice recognition precision can be improved, and the anti-interference capability in the voice acquisition process is improved.

The utility model provides a voice module, turning plate subassembly and range hood, voice module includes shell, pickup piece, connecting piece and has the cover plate of first hole, shell encloses and forms the accommodation cavity, on the shell is provided with the first through hole that communicates the accommodation cavity with the outside, pickup piece at least partially is provided in the first through hole, and the cover plate is provided with the first hole. At least part of the connecting piece is located in the containing cavity, the cover plate is located on the front face of the shell through the connecting piece to shield the pickup piece, and part of the pickup piece is exposed out of the first hole. According to the utility model, on one hand, the pickup function of the pickup member is ensured, and the cover plate can effectively block smudginess and reduce the influence of the smudginess entering the voice module on the pickup member, thereby prolonging the service life of the pickup member; and on the other hand, the cover plate and the shell can be easily connected or separated through the connecting piece, so that quick assembly or disassembly is realized, the assembly and disassembly time is saved, and the pickup piece is convenient to maintain or clean.