83results about How to "Eliminate ambient noise" patented technology

A system and method for capturing, logging and retrieval of face-to-face interactions characterizing walk-in environments. The system comprising a device for capturing and storing one or more face to face interactions captured in the presence of the parties to the interaction, and a database for storing data and metadata information associated with the face-to-face interactions captured.

This invention provides one double microwave sound strength method and device suitable for small mobile communication device to process the input signal x1 and x2 and to adopt wave beam forming technique and use aim sound signal source and noise signal source difference to isolate signals to get the sound signal S(k) and noise signal n(k); using two paths of signals relationship to remove noise part and sound point to get síõ(k) and níõ(k).

A dry powder or aerosol inhaler having attached thereto, in a detachable manner, a sound / vibration sensor adapted to sense sound / vibration in the interval of 100-3000 Hz and there from derive a parameter of a flow in a flow path of the inhaler. The sensor does not interfere with the flow path of the inhaler.

A system and method for capturing, logging and retrieval of face-to-face interactions characterizing walk-in environments. The system comprising a device for capturing and storing one or more face to face interactions captured in the presence of the parties to the interaction, and a database for storing data and metadata information associated with the face-to-face interactions captured.

A dry powder or aerosol inhaler having attached thereto, in a detachable manner, a sound / vibration sensor adapted to sense sound / vibration in the interval of 100-3000 Hz and there from derive a parameter of a flow in a flow path of the inhaler. The sensor does not interfere with the flow path of the inhaler.

The invention discloses an unshielded vehicle-mounted terminal, which comprises an information acquisition module, a data transmission module, a control module and a positioning module, wherein the control module controls the data transmission module, the positioning module and the information acquisition module to work; the information acquisition module is used for acquiring information in the vehicle, around the vehicle and of the vehicle; the data transmission module is used for being connected with other terminals and transmitting data; and the positioning module is used for positioning the vehicle. The unshielded vehicle-mounted terminal is different from the traditional vehicle-mounted terminal equipment in the hardware architecture and product structure, a video display module is not adopted, related video display processing is not required to be performed, a low-specification central processing unit (CPU) processor can be adopted, and the functions such as unshielded navigation, audio and video play, traveling image video recording, vehicle state acquisition, remote control, remote monitoring, hands-free calls and real-time back-up image of the vehicle-mounted terminal are realized at low cost.

An ambient noise removal device includes: a sound signal output section that outputs a sound signal to thereby drive a speaker; a noise detection section that detects an electromotive force produced in the speaker by ambient noise; and a noise cancellation section that generates, on the basis of the electromotive force, a signal of an opposite phase from the ambient noise and adds the generated signal to the sound signal to thereby cancel the ambient noise.

The invention relates to the technical field of electronic communications, in particular to a voice signal processing system. The voice signal processing system with active noise reduction comprises a microphone, an inverse operation unit, a mixing unit and a loudspeaker, wherein the microphone is used for acquiring an environmental noise and converting the acquired environmental noise into an electric signal; the inverse operation unit is connected with an output end of the microphone, and used for performing an inverse operation on the electric signal in order to form an inverse noise signal; the mixing unit is connected with an output end of the inverse operation unit, and used for mixing the inverse noise signal with an audio signal to be played of a playing unit in order to obtain a mixed signal; the loudspeaker is connected with an output end of the mixing unit, and used for playing the mixed signal; and the inverse noise signal in the mixed signal is counteracted with the environmental noise in order to realize noise reduction. Through adoption of the voice signal processing system, the environmental noise can be eliminated actively, and a headset design is simplified, so that the system cost is lowered to the maximum extent while noise reduction is realized.

The invention discloses an audio processing method and device. The audio processing method comprises the steps that audio information is collected through a microphone, and environmental noise in the audio information is filtered out; preset background sounds and the audio information with the environmental noise filtered out are mixed into a path of audio information in real time; the mixed audio information is coded, and therefore the real background sounds are replaced with background sounds set in a user defined mode. Therefore, in the calling process, the background sounds heard by the other person are the background sounds set by the user and are not the background sounds in the real environment, the user can set the background sounds to camouflage the environment where the user is located to achieve the aim of protecting privacy or ridiculing the other person, or in order to remove the crank calls of strangers, the user can set white noise or similar background noise as the background sounds to warn the other person. The environmental noise of the current environment is eliminated, sound recording or calling quality is improved, controllability and interestingness are improved, and the individual needs of the user are met.

Disclosed are an apparatus and a method for voice processing in a mobile communication terminal. A plurality of microphones are used to remove environmental noise at the time of voice communication, so that it is possible to perform high-quality voice communication and video telephony. Moreover, it is possible to perform voice recording even when a user does not open a mobile communication terminal. Furthermore, when voice is recorded or sound is recorded during moving image photographing, a plurality of microphones are effectively utilized to achieve good-quality recording and to perform recording conveniently even when the folder or the slider of the mobile communication terminal is closed. Therefore, it is possible to provide improved convenience in using the mobile communication terminal.

An intelligent voice recognition method, voice recognition apparatus and intelligent computing device are disclosed. An intelligent voice recognition method according to an embodiment of the present invention obtains a microphone detection signal, recognizes a voice of a user from the microphone detection signal and outputs a response related to the voice on the basis of a result of recognition of the voice, wherein the microphone detection signal includes noise, and a microphone detection signal including only the voice obtained by removing the noise from the microphone detection signal is recognized. Accordingly, only a voice of a user can be effectively separated from a microphone detection signal detected through a microphone of the voice recognition apparatus. One or more of the voice recognition devices, intelligent computing devices, and servers of the present invention may include artificial intelligence modules, drones (Unmanned Aerial Vehicles, UAVs), robots, Augmented Reality (AR) devices, and virtual reality (VR) devices, devices related to 5G services, and the like.

The invention relates to the field of mine monitoring and communication and in particular relates to a voiceprint-recognition-based system and method for detecting the uniqueness of a coal mine entering worker. The detection system comprises an entrance / exit control module, a sound-insulated shielding room, a recognition and detection module and an alarm recheck module, wherein the entrance / exit control module is used for controlling that only a detected worker exists at a detection point at the same moment, the worker which can pass through the detection can be released and the worker which can not pass through the detection needs to be subjected to identity rechecking; the sound-insulated shielding room is used for eliminating on-site noise, talking noise with other workers nearby the detected worker and identification card signal interference; the recognition and detection module is used for detecting the number of identification cards, reading voiceprint characteristic information stored in the identification cards, detecting real-time voiceprint characteristic information, and comparing the voiceprint characteristic information so as to obtain a detection result; the alarm recheck module is used for alarming and rechecking the worker who does not pass through the detection. The uniqueness of the coal mine entering worker is detected based on a voiceprint characteristic recognition technology and a worker location identification card technology, and the detection system has the characteristics of high accuracy and low cost.

The invention provides a difference depth feature-based facial expression recognition method and system and belongs to the pattern recognition technical field. The method includes the following stepsthat: face depth features are extracted from a sample face region frame sequence, and neutral expression frames and peak expression frames are extracted from the sample face region frame sequence according to the face depth features; and the face depth features of the neutral expression frames are subtracted from the face depth features of the peak expression frames of the sample image frame sequence, so that difference depth features can be obtained; and the difference depth features of the sample face region frame sequence and corresponding facial expressions are adopted as training input, so that a facial expression classifier can be obtained through training, and the facial expression classifier is adopted to achieve facial expression classification. According to the difference depth feature-based facial expression recognition method and system of the invention, the concept of difference is introduced into the depth features, and the difference depth features are used to representfacial expressions, and therefore, facial expression information can be preserved as much as possible, and at the same time, individual differences and environmental noises can be eliminated; and thedistinguishing capability of used features for the facial expressions is strong, so that the robustness of facial expression recognition can be improved.

The invention discloses a method for building a conducted interference coupling channel multiple linear regression model by near frequency point sampling, and belongs to a conducted electromagnetic interference coupling channel parameter modeling method for building a multiple linear regression model by using the near frequency point sampling under the condition that an interference source is known. Through a frequency segmentation method, the multiple linear regression modeling method is applied to the conducted electromagnetic interference coupling channel parameter modeling with the non-linear relationship characteristics; a multipath observation sample required to modeling in the frequency point is obtained through sampling a series of frequency point characteristics near the sampling observation frequency point. On the precise of no any coupling channel apriori knowledge, the parameter model of the coupling channel can be obtained through extracting the variable relationship of an interference source and a port test signal; the universality is realized.

The invention relates to a cell membrane potential sensor based on a standard complementary metal oxide semiconductor (CMOS) process. The cell membrane potential sensor comprises a metal-oxide-semiconductor (MOS) field-effect tube consisting of a field-effect tube drain region and a field-effect tube source region, the MOS field-effect tube is connected with an oxidation layer, a multi-layer metal grid structure is covered on the oxidation layer and sequentially comprises a connecting layer, a first metal layer, a through hole layer and a second metal layer from bottom to top, and an exposed microelectrode for living cell culture is arranged on the second metal layer. The invention provides the cell membrane potential sensor based on the standard CMOS process, the cell membrane potential sensor realizes the compatibility with the standard CMOS process, and the integration is convenient.

The invention provides a fingerprint induction device and a finger contact detection method. The fingerprint induction device comprises a fingerprint detector and a control device; the fingerprint detector is characterized in that: a plurality of driving lines, a first trace line and a second trace line which are electrically connected to the controller are formed on the substrate, and that the first trace line is arranged between the plurality of driving lines and the second trace line; the controller drives a plurality of driving lines under a working mode and receives the capacitance induction signals corresponding to between the driving lines of the first trace line and the second trace line; under a standby mode, the controller produces a capacitance induction signal between the first trace line and the second trace line and determines whether the finger contacts according to the change of the capacitance induction signal; as a result, the invention can perform detection with relatively big capacitance without driving the plurality of driving lines, can save the electricity, does not need to add the extra trace line to cooperate the energy efficient standby mode, and can perform finger contact detection. The size of the fingerprint detector is not increased.

The embodiment of the invention provides a pickup, pickup equipment and a method for eliminating noise. In the embodiment of the invention, a mixed signal acquired by a first microphone is encoded anddecoded by an audio encoding and decoding chip and then is input into a processor chip; an environmental noise signal acquired by a second microphone is encoded and decoded by the audio encoding anddecoding chip and then is input into the processor chip; and the processor chip reduces the noise of the mixed signal processed by the audio encoding and decoding chip by using the environmental noisesignal processed by the audio encoding and decoding chip, so as to output an available sound signal. The environmental noise signal and the noise signal in the mixed signal acquired by the first microphone are counteracted, so as to acquire the available sound signal; and therefore, an aim of eliminating the environmental noise is fulfilled, and the measurement precision and the stability of thepickup equipment are improved.

The invention relates to a facial expression analysis method and a device based on expression intensity label distribution, and a computer readable medium. The method comprises the following steps: automatically encoding a face image sequence to generate an expression intensity label; and selecting neutral expressions and non-neutral expressions from the expression sequence as input, training theinput data by adopting a convolutional neural network to obtain an expression recognition model, and predicting the category of the expression image and the intensity of the category according to thejoint expression intensity distribution output by the expression recognition model. According to the method, the training data of expression intensity labeling can be obtained, so that the correctnessand robustness of expression recognition are improved.

A voice interaction control guide system and a method of the voice interaction control guide system are a guide function using a voice interaction way to achieve the guide system. A current guide system generally adopts a mouse, a keyboard, a touch interaction of combination of an image and a written language and the like as an interaction way. The above interaction way can not satisfy the needs of different levels of people and different groups of people, such as the blind and children who are illiterate. The voice interaction control guide system carries out corresponding guide functions according to a voice command. A self-adaption voice recognition method and a heuristic prompt method are adopted by the voice interaction control guide system. The voice interaction control guide system stores related words commonly used for guiding in a background database according to property of the guide system, and therefore the system can search in a corresponding database when a user sends a voice command, and matching speed is improved. If corresponding matching words and sentences are not searched, the system responds and asks whether corresponding operation is needed to be done by the adoption of voice, and then the user just answers 'yes' or 'no', the system enters the next guide, and a humanized guide function is achieved.

The invention relates to a Doppler flow rate meter data processing method and system, a medium and computer equipment. The method comprises a step of acquiring frequency spectrum data of a water areameasured by a Doppler flow rate meter, a step of judging the type of a flow field in the water area according to spectrogram characteristics of the frequency spectrum data of the water area, and a step of selecting a corresponding flow rate signal identification algorithm to calculate a flow rate value according to the type of the flow field. Aiming at different spectrogram characteristics of frequency spectrum data obtained in different scenes and different flow places, the type of the flow field in the water area detected by the flow rate meter at a moment is effectively recognized, different flow rate signal recognition algorithms are automatically switched, an accurate and stable flow rate value is obtained, the repeated measurement is not needed, and the power consumption is low.

The invention discloses a weak-reflection Bragg grating accelerometer and a sensing method thereof. First, second and third gratings which are distributed at equal intervals are graved in a polarization-maintaining sensing optical fiber; two pulses which are mutually independent in a time domain return after sequentially going through the first, second and third gratings to generate two interference pulses carrying outside vibration information, phase of the interference pulses carrying the outside vibration information is demodulated and then reduced so as to acquire outside vibration information inhibiting common-mode noise, and phase difference is proportional to outside vibration signals. Noise suppression level is elevated, common-mode noise introduced by a laser device and an acoustic optical modulator can be eliminated, environment noise introduced to the accelerometer itself by perturbation of outside environment can be eliminated, phase sensitivity is improved, transverse crosstalk is inhibited, cost of a sensing system is saved while size of the same is reduced, and this advantage is obviously demonstrated especially in a quasi-distributed multi-motif sensing network.

A bowel sound analysis method and system utilizes an audio collection apparatus to continuously monitor an abdominal cavity of an examinee and collect a bowel sound signal of an intestinal tract inside the abdominal cavity, converts the bowel sound signal into a digital signal, then removes noise from the digital signal by using higher-order statistics (HOS) and captures a high-complexity feature from the digital signal by using fractal dimension algorithm, and subsequently defines the high-complexity feature as an intestinal motility signal. The refore, according to the intestinal motility signal, a time point when intestinal motility occurs in the intestinal tract inside the abdominal cavity of the examinee can be learned, and bowel sound features such as the amplitude, frequency, and period of the intestinal motility signal may further be analyzed to determine a health condition of the intestinal tract.

The invention discloses an uncoiler automatic coiling control system based on image processing, and the system comprises an image collection module which is connected with an image processing module and a controller module through the Ethernet. The invention further discloses a method for performing target detection by using the uncoiler automatic coiling control system based on image processing,background noise and environmental noise of images are eliminated by adopting multiple filtering modes, the detection precision is improved, an error judgment module is designed, the detection reliability is calculated, and the safety and reliability in industrial production are ensured.

The embodiment of the invention provides a pillow and pillow combination capable of eliminating noise. The pillow comprises a noise collector, a loudspeaker and a noise eliminating device which are arranged in the pillow, and the noise collector and the loudspeaker are in communicated connection with the noise eliminating device separately; the noise collector is used for collecting noise signals sent by a noise source and sending the noise signals to the noise eliminating device; the noise eliminating device is used for receiving the noise signals sent by the noise collector, generating noise eliminating signals corresponding to the noise signals according to the noise signals and sending the noise eliminating signals to the loudspeaker; the loudspeaker is used for receiving the noise eliminating signals sent by the noise eliminating device and radiating the noise eliminating signals in air to counteract the noise signals. According to the pillow and pillow combination capable of eliminating the noise, the environmental noise can be eliminated, and the user sleeping quality can be improved.

The present invention comprises an apparatus for ascertaining a personal tinnitus frequency consisting of an audio signal generator unit and an associated audio signal earphone, which apparatus provides a first audio signal whose frequency is continuously adjustable. The invention is characterized in that compensating means are provided in order to eliminate ambient noise through active noise reduction.

A system for interactive reality activity augmentation includes a sensing unit, digital projectors and a server. The sensing unit illuminates activity surfaces and captures one of series of real-time images and co-ordinates information of the activity surfaces. The server includes a memory and a processor. The processor receives and processes one of the series of received real-time images and the co-ordinates information to detect presence and tracks trajectories and calculate individual co-ordinates of physical activity objects, virtual objects and users along the activity surfaces to detect interaction information, and feeds the calculated co-ordinates and the interaction information to a scheduler through an application program interface to manipulate one or more contents running inside the scheduler in response to the co-ordinates information and the interaction information. The digital projectors project the one or more manipulated contents over the activity surfaces to provide the interactive reality augmented activity experience.

The invention provides a hearing testing system and method to improve the accuracy of hearing testing and lower the operation difficulty of hearing testing. The system comprises noise-canceling headphones (1) with a self-calibration function, an intelligent terminal (2) and hearing testing application software (3). The hearing testing method based on the hearing testing system comprises the steps that 1, self-calibration is conducted on the noise-canceling headphones with the self-calibration function; 2, the noise-canceling headphones with the self-calibration function is connected with the intelligent terminal, and a noise-canceling function is started; 3, hearing testing is conducted by means of the hearing testing application software located in the intelligent terminal. According to the hearing testing system and method, implementation and operation are easy, and the testing result is accurate and credible.

There is provided an object detection device including a light source, an avalanche diode, a first counter, a second counter and a processor. The light source emits light within a first interval, and is turned off within a second interval, a third interval and a fourth interval. The avalanche diode detects photons corresponding to the first interval, second, third and fourth intervals to trigger avalanche events. The first counter performs up-counting on the avalanche events in the first interval and performs down-counting on the avalanche events in the third interval to generate a first count value. The second counter performs up-counting on the avalanche events in the second interval and performs down-counting on the avalanche events in the fourth interval to generate a second count value. The processor calculates a time-of-flight according to the first count value and the second count value.

A device and a method for filtering out noise from speech of caller are disclosed. The method is applied to the device, includes: inputting a speech sound of a caller; converting the speech sound to digital signals by an analyzing-to-digital converting unit; analyzing the digital signals to identify a pure speech of the caller and filtering out an extraneous noise thus obtaining pure speech signals of the caller; encoding the pure speech signals by a coder and decoder unit, and submitting the encoded speech signals to the receiver.