525 results about "External noise" patented technology

The invention relates to a transformer multi-physics field coupling numerical simulation method and device and a storage medium. The method comprises the following steps: performing coupling calculation on a target working condition by using an electromagnetic, heat, flow and force coupling calculation model of a transformer to obtain transformer loss data and temperature rise data; mapping the temperature rise data to a transformer electromagnetic and force coupling calculation model by using a grid node data mapping method to serve as an initial value, and performing coupling calculation on a target working condition to obtain dynamic data of an internal structural member of the transformer; performing fast Fourier transform on the dynamic data in a preset frequency range to obtain dynamic data under each frequency; and mapping the temperature rise data and the dynamic data under each frequency to a transformer sound and force coupling calculation model as initial values, and carrying out coupling calculation on a target working condition to obtain transformer oil tank vibration and external noise data. Compared with the prior art, the method has the advantages that the calculation amount is effectively reduced, and the accuracy of predicting the internal state of the transformer is improved.

Disclosed are a filter and a noise-reducing earplug, the noise-reducing earplug includes a filter, and the filter includes a shell, a flexible sound filtering block, and an adjustment pressing plate, wherein the shell is provided with a mounting recess and a sound outlet hole communicating with the mounting recess, the adjustment pressing plate is mounted in the mounting recess and forms a compression cavity with the mounting recess, the adjustment pressing plate is provided with sound inlet holes, the flexible sound filtering block is mounted in the compression cavity, the adjustment pressing plate contacts an end of the flexible sound filtering block away from the sound outlet hole, and the adjustment pressing plate is configured to adjust a volume of the compression cavity. The noise-reducing earplug provided in the present disclosure may adjust the noise-reducing effect; and a user may selectively filter external noise to different degrees as needed.

The invention discloses an SER6 Pro antenna with adaptive beam forming capability, comprising: a generation module for generating an initial beam directional diagram based on spatial-temporal correlation analysis of multipath signals; the calculation module is used for updating beam direction parameters in real time by using a Doppler frequency shift estimation algorithm so as to adapt to Doppler frequency shift change of a high-speed mobile user; the suppression module is used for carrying out noise suppression on the beam signal in a frequency domain by adopting an adaptive filter; and the optimization module is used for dynamically adjusting the beam forming matrix and optimizing the beam direction and intensity through an iterative optimization algorithm. The method has the beneficial effects that the problem of inaccurate beam direction identification caused by excessive reflected and scattered signals in a complex multi-path propagation environment is solved; the problem of unstable beam tracking caused by significant change of Doppler frequency shift in a high-speed mobile user scene is solved; and the problem that the signal-to-noise ratio is reduced due to the strong external noise source in a high-interference environment is solved.

The invention discloses an MFL signal reconstruction method based on channel mutual fusion and a global attention mechanism, relates to the technical field of nondestructive testing, and particularly discloses a magnetic flux leakage signal MFL reconstruction method based on a neural network. In order to solve the problem of interference of external noise on defect characteristic signals, a signal channel recombination, shuffling and aggregation optimization method based on CycleGAN is adopted, a GSoP optimization covariance matrix is introduced, through PSNR and LPIPS index comparison, the reconstruction precision of the MFL defect signals can be greatly improved, and the reconstruction precision of the MFL defect signals is improved. And the data processing precision and speed are improved (the time cost and the space cost of data processing are reduced), so that the defect detection efficiency is improved.

The invention relates to the technical field of signal calibration, in particular to a main radial probe signal calibration method for a medical cyclotron, and the method comprises the steps: collecting probe current data, and constructing a probe current data sequence; further obtaining a time domain amplitude fluctuation degree; calculating the time domain baseline stability degree of each period; obtaining the time domain noise interference degree of each probe position; obtaining the frequency domain bandwidth difference of each probe position; obtaining the frequency domain noise interference degree of each probe position; and acquiring an external noise interference degree based on the time domain noise interference degree and the frequency domain noise interference degree, and calibrating the main radial probe signal. According to the invention, the actual beam current intensity can be evaluated more accurately.

The invention discloses multi-frequency array eddy current detection equipment and a method thereof, belongs to the technical field of eddy current detection, and aims to find out an outlier in an extracted phase difference according to fluctuation intensity and fluctuation range of the phase difference and calibrate the outlier so as to efficiently prevent the defect of inaccurate phase difference detection caused by external noise fluctuation interference. According to the relation between the eddy current detection numerical value and the phase difference of the transverse pipeline, the similarity between the eddy current detection condition of the numerical cluster corresponding to each previous numerical value point and the eddy current detection condition of the numerical cluster where the outlier is located, and the action amplitude of each numerical value point in the numerical cluster on the outlier, finding out a previous numerical value similar to the outlier; and the outliers are calibrated, so that the precision of the phase difference can be improved, and the calculation of the wall thickness of the pipeline according to the phase difference is improved until the precision of detecting the local micro crack defect of the pipeline is achieved.

A measurement system for determining a noise figure of a device under test is described. The measurement system determines a first total error power based on the output signal by a first noise canceling technique, wherein the first total error power includes systematic errors originating in the device under test and in the measurement system. The measurement system determines a second total error power based on the output signal by a different, second noise canceling technique. The second total error power includes systematic errors originating in the measurement system. The second total error power further includes noise originating in the device under test and in the measurement system outside of the measurement instrument. The measurement system is configured to subtract the first total error power from the second total error power, thereby obtaining an external noise power.

The invention relates to the technical field of information security and random number generation, and particularly discloses a method and a system for generating a random number by utilizing mobile phone holding micro-vibration and touch pressure, and the random number is generated by collecting physical behavior characteristics when a user holds a mobile terminal within preset time and combining sensor signal processing and a randomness extraction algorithm. The random number generation does not need external hardware support, the randomness is higher, the operation is simple and convenient, and the problems of insufficient randomness and complex operation caused by dependence on an external noise source or specific input in the prior art are solved. According to the method, the dependence on the external environment or additional hardware is avoided, the security, unpredictability and practicability of random number generation are remarkably improved through a multi-level security protection mechanism, reliable technical support is provided for the field of modern information security, and the method has the advantages of being easy to implement, low in cost and high in security and has wide application prospects. The method is suitable for popularization and application in multiple fields of mobile payment, Internet of Things, games and the like.

A communications headset includes one or more directional speakers mounted in left and right temple housings positioned near a wearer's ears, with acoustic channels within the temple housings configured to direct sound waves toward the ear canals. The temple housings form an open-ear configuration, allowing external noises to reach the ears unobstructed. Each directional speaker is acoustically coupled to a contoured channel that delivers focused audio to the wearer's ear with minimal sound wave leakage. An in-ear speaker, magnetically secured and tethered to the temple housing, provides optional noise-isolated listening. A user-controlled selection switch allows a wearer to toggle between open-ear and in-ear modes. The headset also includes a boom microphone, volume controls, and USB charging and communication ports.

The invention relates to the field of optical fiber sensing, in particular to a DAS-based audio data acquisition method and system, electronic equipment and a storage medium. Acquiring real-time audio data and real-time energy of the to-be-measured optical cable at the current sampling time point; updating the reference noise threshold value through the reference energy in the real-time audio data; acquiring a starting point of an effective signal in the real-time audio data according to the real-time energy and the updated reference noise threshold value, and amplifying the real-time audio data collected behind the starting point; and taking the amplified real-time audio data as effective audio data. Compared with the prior art, the method has the advantages that the reference noise threshold value containing the system noise is obtained, the reference energy containing the external noise is used for updating the reference noise threshold value, then the effective signal can be obtained based on the updated reference noise threshold value and the real-time energy, calculation is simple, and no complex algorithm or extra hardware is needed.

The invention discloses a wind noise simulation method, device and equipment based on a commercial vehicle and a medium. The method comprises the following steps: constructing a whole vehicle grid model matched with a target commercial vehicle to be subjected to noise simulation; importing the whole vehicle grid model into a pre-constructed blank simulation model framework to obtain a basic simulation model; setting free attribute parameters and material attribute parameters of the basic simulation model according to regional noise characteristics and material attributes of each key region in the whole vehicle grid model to obtain an external noise simulation model and an internal vocal cavity simulation model; and operating the external noise simulation model and the internal vocal cavity simulation model to obtain a wind noise simulation result of the target commercial vehicle. According to the embodiment of the invention, a comprehensive wind noise simulation result is provided, a powerful basis is provided for vehicle noise control and optimization, analysts are helped to identify noise sources and evaluate the effect of noise control measures, and vehicle design is optimized to reduce the noise level.

The invention discloses a filter parameter optimization method. The method is characterized in that a first feed-forward microphone, a second feed-forward microphone and a virtual microphone are respectively arranged; the first feed-forward microphone acquires a first signal, the second feed-forward microphone acquires a second signal, and the virtual microphone acquires an expected signal; adding and mixing the first signal and the second signal after passing through the first feed-forward controller and the second feed-forward controller to obtain an adding and mixing signal; carrying out difference mixing on the addition mixed signal and the expected signal through a secondary path to obtain a residual signal; and respectively adjusting filter parameters of the first feed-forward controller and the second feed-forward controller through the residual information. Parameters of the active noise reduction filter can be adjusted according to external noise conditions and user requirements, so that good noise reduction effect and audio quality are realized.

The utility model discloses a butterfly-shaped laser test fixture with a shield, which comprises a first shell, and a second shell is arranged at the bottom of the first shell. The first shell is provided with a clamping unit which is used for guaranteeing that pins at all positions are uniformly stressed in the testing process of the butterfly-shaped laser, a temperature stabilizing unit which is used for cooling the butterfly-shaped laser in the testing process, and a protection structure which is used for shielding the influence of external noise and radiation on the butterfly-shaped laser in the testing process. Pins of the butterfly-shaped laser can be uniformly pressed and communicated with the control circuit board through the clamping unit, and the butterfly-shaped laser is wrapped with the protection structure, so that the influence of external noise and radiation on the butterfly-shaped laser is shielded in the testing process; the temperature stabilizing unit is used for dissipating heat of the tested butterfly-shaped laser; the accuracy of test data of the butterfly-shaped laser in the test process is greatly improved, and the technical problem that the accuracy of test data of an existing butterfly-shaped laser in the test process is still low is solved.

A magnetic shield reduces external noise in a chamber including a target and at least one electromagnet for copper physical vapor deposition (PVD). The shield may have a thickness in a range from approximately 0.1 mm to approximately 10 mm to provide sufficient protection from radio frequency and other electromagnetic signals. As a result, copper atoms in the chamber undergo less re-direction from external noise. Additionally, even when hardware failure occurs during PVD (e.g., an electromagnet malfunctions, a wafer stage is not level, and / or a flow optimizer induces too much shift, among other examples), the copper atoms are less susceptible to small re-directions from external noise. As a result, back end of line (BEOL) and / or middle end of line (MEOL) conductive structures are formed in a more uniform manner, which increases conductivity and improves lifetime of an electronic device including the BEOL and / or MEOL conductive structures.

The invention discloses a two-channel graph convolution contrast learning recommendation method and system fused with multi-dimensional scoring, and belongs to the technical field of deep learning. The multi-dimensional scoring system is a new scoring system which is developed on each large online platform in recent years, comprises total scores of the user on the items and scores of the user on the items on different attributes, and can comprehensively and meticulously reflect user preferences. However, most of the existing personalized recommendation algorithms are oriented to a single-dimensional scoring system only containing a total score, and ignore fine-grained preferences of a user in different dimensions and a complex relationship between the fine-grained preferences and the total score. On the basis, a two-channel graph convolution contrast learning recommendation method fusing multi-dimensional scores is provided, a graph convolution neural network and an attention mechanism are comprehensively utilized to learn coarse and fine granularity characterization of users and items, and a high-order complex relation between each attribute and a total score is modeled through two-channel contrast learning. While the user attribute preference and the representation vector are optimized, the introduction of external noise is effectively avoided, and a high-quality personalized recommendation result is provided for the user.

1. The name of the product of this design: earplugs. 2. Purpose of the product of this design: earplugs used to reduce external noise. 3. The design key point of this appearance design product lies in the shape. 4. The picture or photo that best illustrates the design points: combination status diagram.

This present invention relates to a single body air treatment system for recreational vehicles and is a combination of components working in a scheme so to provide heating, cooling, dehumidification, ventilation, air-purification, heavy-duty drying, into a single body. The system comprising of a cooling air plant having a refrigerating cycle (105,107,108,109) and air heat exchangers acting with electricity and a heating plant having an incorporated fuel / Natural-Gas burner (118,119) acting as air furnace. The system operates heating and cooling simultaneously with in the same air-flow and provides air purification through an antibacterial electrical device. The system also integrates extra pipes to produce extra hot-water in winter while the system operates in heating. The system is integrated with a dedicated resonator chamber to minimize the external noise produce by extraction blowers.

The utility model relates to the technical field of calcium silicate boards, in particular to a calcium silicate board waterproof structure which comprises a structure body, a first calcium silicate board layer is arranged at the top of the structure body, a colored surface is arranged at the bottom of the first calcium silicate board layer, an anti-corrosion layer is arranged in the structure body, a drying layer is arranged in the anti-corrosion layer, and a fireproof frame is arranged in the drying layer. A silencing structure is arranged in the fireproof frame and mainly comprises a silencing layer, a sound transmission hole, a silencing opening, silencing cotton and a silencing medium. Through the arrangement of the sound transmission holes, the silencing openings, the silencing cotton and the silencing medium, noise transmission can be reduced through the silencing layer, external noise is absorbed through the sound transmission holes, the absorbed noise is transmitted to the interior of the sound absorption medium, and the absorbed noise is circularly transmitted and reduced through the silencing holes in the sound absorption medium; and through the silencing cotton, the transmission of noise can be reduced, the volume of the noise is effectively reduced, and the sound insulation effect of the sound reduction layer is ensured.

A magnetic shield reduces external noise in a chamber including a target and at least one electromagnet for copper physical vapor deposition (PVD). The shield may have a thickness in a range from approximately 0.1 mm to approximately 10 mm to provide sufficient protection from radio frequency and other electromagnetic signals. As a result, copper atoms in the chamber undergo less re-direction from external noise. Additionally, even when hardware failure occurs during PVD (e.g., an electromagnet malfunctions, a wafer stage is not level, and / or a flow optimizer induces too much shift, among other examples), the copper atoms are less susceptible to small re-directions from external noise. As a result, back end of line (BEOL) and / or middle end of line (MEOL) conductive structures are formed in a more uniform manner, which increases conductivity and improves lifetime of an electronic device including the BEOL and / or MEOL conductive structures.

The invention discloses a whole vehicle aerodynamic noise prediction method, device and equipment, a storage medium and a product. The method comprises the following steps: determining an aerodynamic noise prediction result of a to-be-predicted commercial vehicle under an external noise source; the noise source outside the automobile comprises an air conditioner air outlet and a pressure release valve. Performing aerodynamic noise transmission analysis on a sealing system and a sound insulation system of the commercial vehicle to be predicted according to the noise prediction result of the noise source outside the vehicle, and determining an aerodynamic noise transmission result; according to the aerodynamic noise transmission result and the aerodynamic noise prediction result of the air conditioner air outlet and the pressure release valve, aerodynamic noise transmission analysis is conducted on a cockpit sound absorption system of the to-be-predicted commercial vehicle, and the aerodynamic noise response prediction result of the whole to-be-predicted commercial vehicle at the human ears is determined. According to the technical scheme provided by the embodiment of the invention, the accurate prediction, simulation or simulation of the aerodynamic noise response of the commercial vehicle to the human ears of the driver in the vehicle driving process is improved.

The utility model relates to the field of semiconductors, and discloses a semiconductor device, which comprises a box body, the box body comprises a box chamber and a box wall enclosing the box chamber, the box wall comprises a first noise reduction structure, one side of the box wall is provided with a chip conveying port, a semiconductor detection mechanism is fixedly arranged in the box chamber, and the chip conveying port is arranged opposite to a material port; the door mechanism is slidably connected with the box wall, and the door mechanism comprises a second noise reduction structure. In order to solve the problem that the operation precision of the semiconductor equipment is reduced due to the influence of external noise, according to the semiconductor equipment of the embodiment of the invention, the semiconductor detection mechanism is arranged in the box chamber of the box body, and the box wall is provided with the wafer conveying port to ensure the normal work of the semiconductor detection mechanism; the door mechanism with the second noise reduction structure performs sealing and noise reduction on the wafer conveying opening, sound insulation and noise reduction on the semiconductor detection mechanism are improved through the box wall with the first noise reduction structure and the door mechanism, the influence of external noise on the semiconductor detection mechanism is reduced, and the working precision of the semiconductor detection mechanism is improved.

The invention relates to an abnormal sound detection method, which comprises the following steps of: installing an abnormal sound detection device at a to-be-detected position of an automobile part, and detecting a vibration signal at an abnormal sound occurrence moment by using the abnormal sound detection device; and analyzing the vibration signal, determining an abnormal sound frequency band, performing noise reduction on the abnormal sound frequency band, and analyzing the to-be-detected position according to the vibration signal after noise reduction. The abnormal sound detection device is installed at the to-be-detected position of the automobile part, the abnormal sound detection device is used for detecting the vibration signal at the abnormal sound occurrence moment, then the vibration signal is analyzed, the abnormal sound frequency band is determined, the detection range of the vibration signal is reduced, noise reduction is conducted on the abnormal sound frequency band, the signal-to-noise ratio of the signal of the abnormal sound frequency band is increased, and the abnormal sound detection accuracy is improved. The interference of external noise is reduced, the feature information of the vibration signal is improved, and the technical problem that the fault position cannot be accurately judged due to the fact that the abnormal sound signal detected by the abnormal sound stethoscope in the vehicle driving process is interfered by other irrelevant signals in the related technology is solved.

The invention relates to a manufacturing method of a high-heat-insulation and sound-insulation type PVB (Polyvinyl Butyral) intermediate film for an automobile large skylight, and belongs to the technical field of laminated glass material preparation. A common PVB resin layer A, a high-damping-coefficient sound-insulation resin layer B, a heat-insulation composite resin layer C, a high-damping-coefficient sound-insulation resin layer B and a common PVB resin layer A are prepared into the high-heat-insulation sound-insulation type PVB intermediate film in sequence by adopting a five-layer composite co-extrusion process, a multi-layer structure is adopted, and particularly, the high-damping-coefficient sound-insulation resin layer B is used, so that the propagation of sound waves can be effectively reduced, and the sound-insulation performance of the PVB intermediate film is improved. The large skylight of the automobile can effectively isolate external noise and improve the riding comfort in the automobile during high-speed driving or urban traffic, vanadium dioxide and graphene are compounded through the C-layer heat insulation composite resin, graphene has a large specific surface area, the dispersity of vanadium dioxide can be improved, and the heat insulation performance of the automobile is improved. Vanadium dioxide can improve the reflectivity of near-infrared light when the environment temperature rises.

The invention discloses an injection molding type assembly test probe and a production method thereof, the injection molding type assembly test probe comprises a tube body, two ends of the tube body are provided with an upper probe body and a lower probe body, one end of the tube body close to the upper probe body is provided with a flange plate, the outer wall of the upper probe body is provided with a limiting ring, and when the upper probe body is sleeved in the tube body, the limiting ring is arranged on the flange plate. A first fastening piece and a second fastening piece which are made of insulation materials are arranged between the pipe body and the upper needle body and between the pipe body and the lower needle body respectively, and the first fastening piece is used for tightly matching the flange plate and the limiting ring in the axial direction of the pipe body. And the second fastener is used for enabling the port of the tube body to be in interference fit with the outer wall of the lower needle body in the radial direction of the tube body. Interference of external noise on high-frequency signals is blocked, excellent isolation performance is achieved, in addition, due to the fact that injection molding is adopted, welding spots cannot be formed on the pipe body, impedance stability in the signal transmission process can be better guaranteed, reflection loss is reduced, and signal integrity is improved.

The invention relates to a fabric friction sound testing device which comprises a noise reduction box, an auxiliary friction table, an objective table, a weight increasing weight, a drafting device and a traction rope. One side of the noise reduction box extends outwards to form a noise elimination channel; the auxiliary friction table is fixed to the inner bottom of the noise reduction box. The objective table is placed on the auxiliary friction table; the weight increasing weight is used for being placed in the objective table. The drafting device is arranged outside the noise reduction box; the traction rope penetrates through the silencing channel, one end of the traction rope is connected with the objective table, and the other end of the traction rope is connected with the drafting device; and the sound acquisition device is fixed in the noise reduction box. The traction device is arranged outside the noise reduction box, the bottom noise can be effectively reduced fundamentally, the accuracy of a test result is improved, one side of the noise reduction box extends outwards to form the noise elimination channel, the traction rope penetrates through the noise elimination channel, and the noise elimination channel is arranged, so that the noise reduction effect is improved. The noise reduction box can effectively reduce the influence of holes formed in the path of the traction rope on the noise reduction box on the sound insulation effect, avoids external noise interference, and ensures the accuracy of a test result.

The invention discloses a fault diagnosis method based on multi-source data fusion and a deep optimization network, and belongs to the technical field of wind turbine generator bearing fault diagnosis, and the method comprises the steps: obtaining multi-source sensor data of a wind turbine generator under different working conditions, and carrying out the preprocessing of the collected multi-source sensor data; designing a multi-source data feature fusion algorithm based on correlation variance contribution, fusing the preprocessed multi-source sensor data, and extracting a fuzzy entropy value from the fused multi-source sensor data as a feature vector input into the intelligent fault diagnosis model; constructing an intelligent fault diagnosis model DBE based on the optimized deep belief network, and training the DBE by adopting a greedy learning method and a hippocampus optimization algorithm; and performing fault diagnosis based on the trained DBE. According to the method, the problems of signal abnormal value and data missing and fault diagnosis reliability under the influence of variable working conditions and external noise interference of the wind turbine generator are solved, and the fault diagnosis accuracy of the wind turbine generator is improved.

The invention provides an anti-interference shielding method for a capacitor outgoing line, and relates to the technical field of signal anti-interference shielding, and the method comprises the steps: installing a filter group which comprises a multi-stage filtering device and a self-adaptive filter; identifying an input signal by using a spectrum analyzer, and analyzing and determining multi-source interference characteristics; determining multi-stage working parameters of a multi-stage filtering device according to the multi-source interference frequency, the multi-source interference intensity and the multi-source duration matching, and optimizing to obtain optimal working parameters of an adaptive filter; and controlling the filter bank to perform anti-interference shielding according to the multi-stage working parameters and the optimal working parameters. The technical problems that an existing anti-interference shielding method of the capacitor outgoing line is difficult to effectively deal with a complex interference source and poor in anti-interference performance can be solved, interference signals of different frequency bands can be effectively restrained through combination of multi-stage filtering and the self-adaptive filter, and the anti-interference performance of the capacitor outgoing line is improved. The influence of external noise on a capacitor outgoing line power supply port is reduced, and the anti-interference capability is improved.

The invention discloses a noise and vibration energy acquisition system and a dynamic monitoring method thereof, and belongs to the technical field of energy conversion, the noise and vibration energy acquisition system comprises a sound input unit, a monitoring shell, an energy conversion unit, a vibration sensing unit and a storage control unit; an acoustic resonance cavity is arranged in the monitoring shell, and a vibration sensing unit, an energy conversion unit and a sound input unit are arranged in the acoustic resonance cavity from bottom to top; the vibration sensing unit and the energy conversion unit are electrically connected with the storage control unit respectively; the vibration sensing unit is used for acquiring an external vibration signal and generating an alternating current signal, the energy conversion unit is used for acquiring an external noise signal and generating an alternating current signal, and the storage control unit is used for respectively acquiring the alternating current signals generated by the vibration sensing unit and the energy conversion unit; the generated electric energy is supplied to an external load and the intensity of external vibration and noise is obtained. Synchronous acquisition of noise and vibration energy can be realized, and real-time dynamic monitoring of external noise and vibration intensity can be realized.

The utility model relates to the technical field of automobiles, in particular to a top cover front cross beam assembly and an automobile, and aims to solve the problem that the NVH (Noise Vibration and Harshness) performance of an existing automobile is difficult to effectively optimize while the structural strength of an automobile body is ensured. Therefore, the top cover front cross beam assembly comprises an outer plate, an inner plate and a reinforcing plate arranged between the outer plate and the inner plate, the reinforcing plate comprises a middle area and end areas located on the two sides of the middle area in the length direction of the reinforcing plate, and a first cavity and a second cavity which are separated are formed between the end areas and the inner plate. The first cavity and the second cavity are distributed in the width direction of the reinforcing plate, and a third cavity is formed between the middle area and the inner plate. Therefore, the structural strength and the collision safety of the whole vehicle body can be remarkably improved, the propagation path of sound waves in the cross beam can be reduced, the resonant frequency is reduced, external noise and vibration are effectively prevented from being transmitted into a vehicle cabin, and the NVH performance is remarkably optimized.

A method performed by a wearable audio output device worn by a user is provided for controlling external noise attenuated by wearable audio output device. A speech is detected from a user wearing the wearable audio output device, wherein the audio output device has active noise reduction turned on. It is determined, based on the detecting, that the user desires to speak to a subject in the vicinity of the user. In response to the determining, a level of noise reduction is reduced to enable the user to hear sounds external to the audio output device. It is determined that the user desires to speak to the subject by detecting at least one condition of a plurality of conditions.