Active noise reduction with pulse detection and suppression

Abstract

In one embodiment, an apparatus includes a noise reduction headset including one or more microphones configured to generate an input signal and an acoustic transducer; a controller comprising one or more processing devices, the controller configured to: process the input signal by one or more noise reduction filters to generate a noise-reduced signal; comparing the input signal with an estimate of ambient noise to determine whether the energy of the input signal is greater than the estimate of ambient noise, where a change in the noise-reduced signal is suppressed if the energy of the input signal is greater than the estimate of ambient noise by a predetermined amount; and generating an output signal comprising, at least in part, the noise-reduced signal, where the acoustic transducer is configured to generate an acoustic output from the output signal.

Description

Background Art

[0001] The present invention generally relates to acoustic devices such as headphones, which may include active noise reduction (ANR) capabilities that block at least a portion of ambient noise from reaching a user's ear, and more particularly to acoustic devices having ANR capabilities that detect and suppress ANR responses to impulse sounds. Summary of the Invention

[0002] Two or more features described in this disclosure, including those described in the Summary of the Invention section, may be combined to form specific embodiments not specifically described herein.

[0003] According to one aspect, a device includes: a noise-canceling headphone including one or more microphones and a sound transducer, the one or more microphones being configured to generate an input signal based on captured ambient sound; a controller including one or more processing devices, the controller being configured to: process the input signal through one or more noise-canceling filters to generate a noise-canceling signal, wherein the noise-canceling signal is configured to reduce the effect of the input signal; compare the input signal with an estimate of ambient noise to determine whether the energy of the input signal is greater than the estimate of ambient noise, wherein if the energy of the input signal is greater than the estimate of ambient noise by a predetermined amount, suppress changes in the noise-canceling signal; and generate an output signal, the output signal at least partially including the noise-canceling signal, wherein the sound transducer is configured to produce a sound output based on the output signal.

[0004] In an example, the output signal is a weighted combination of the noise-canceling signal and a pass-through signal.

[0005] In an example, comparing the input signal with an estimate of ambient noise includes: comparing the energy of the input signal with an ambient noise signal generated by a first low-pass filter, wherein the first low-pass filter is configured such that the ambient noise signal is an estimate of the ambient noise present in the captured ambient sound.

[0006] In an example, the ambient noise signal is delayed in time relative to the input signal.

[0007] In an example, the energy of the input signal is determined by the output of a second low-pass filter, wherein the second low-pass filter has a greater effect on smoothing the input signal than the first low-pass filter.

[0008] In an example, comparing the energy of the input signal further includes: determining whether a difference between the output of the second low-pass filter and the ambient noise signal satisfies a threshold condition.

[0009] In an example, comparing the energy of the input signal further includes: determining whether a ratio between the output of the second low-pass filter and the ambient noise signal satisfies a threshold condition.

[0010] In an example, suppressing the noise reduction signal includes: temporarily stopping to increase the amplitude of the noise reduction signal.

[0011] In an example, temporarily stopping to increase the amplitude of the noise reduction signal includes: temporarily stopping to adjust a variable gain filter in a direct-through processing chain that generates a direct-through signal, where the output signal is a weighted combination of the noise reduction signal and the direct-through signal.

[0012] In an example, suppressing the noise reduction signal includes: adjusting a rate at which the noise reduction signal is adjusted in response to the input signal.

[0013] According to another aspect, one or more non-transitory machine-readable storage devices have computer-readable instructions encoded thereon for causing one or more processing devices to perform a method that includes the steps of: receiving an input signal from one or more microphones, the input signal being based on captured ambient sound; processing the input signal through one or more noise reduction filters to generate a noise reduction signal, where the noise reduction signal is configured to reduce an impact of the input signal; comparing the input signal with an estimate of ambient noise to determine whether an energy of the input signal is greater than the estimate of the ambient noise, where if the energy of the input signal is greater than the estimate of the ambient noise by a predetermined amount, suppressing a change in the noise reduction signal; and generating an output signal for an acoustic transducer, the output signal at least partially including the noise reduction signal, such that the acoustic transducer produces an acoustic output based on the output signal.

[0014] In an example, the output signal is a weighted combination of the noise reduction signal and a direct-through signal.

[0015] In an example, comparing the input signal with the estimate of ambient noise includes: comparing an energy of the input signal with an ambient noise signal generated by a first low-pass filter, where the first low-pass filter is configured such that the ambient noise signal is an estimate of the ambient noise present in the captured ambient sound.

[0016] In an example, the ambient noise signal is delayed in time relative to the input signal.

[0017] In an example, the energy of the input signal is determined by an output of a second low-pass filter, where the second low-pass filter has a greater impact on smoothing the input signal than the first low-pass filter.

[0018] In an example, comparing the energy of the input signal further includes: determining whether a difference between the output of the second low-pass filter and the ambient noise signal satisfies a threshold condition.

[0019] In an example, comparing the energy of the input signal further includes: determining whether a ratio between the output of the second low-pass filter and the ambient noise signal satisfies a threshold condition.

[0020] In an example, suppressing the noise reduction signal includes: temporarily stopping to increase the amplitude of the noise reduction signal.

[0021] One or more non-transitory machine-readable storage devices according to claim 18, wherein temporarily stopping to increase the amplitude of the noise reduction signal includes: temporarily stopping to adjust a variable gain filter in a direct-through processing chain that generates a direct-through signal, wherein the output signal is a weighted combination of the noise reduction signal and the direct-through signal.

[0022] One or more non-transitory machine-readable storage devices according to claim 11, wherein suppressing the noise reduction signal includes: adjusting a rate at which the noise reduction signal is adjusted in response to the input signal.

[0023] According to another aspect, a method includes the steps of: receiving an input signal that represents audio captured by a microphone of an active noise reduction (ANR) headset; processing a portion of the input signal by one or more processing devices to determine a noise level in the input signal; determining that the noise level satisfies a first threshold condition; comparing the input signal with an estimate of ambient noise to determine whether the energy of the input signal is greater than the energy of the estimate of the ambient noise by a predetermined amount, and in response to determining that the noise level satisfies the first threshold condition and the energy of the input signal is not greater than the energy of the estimate of the ambient noise by the predetermined amount, generating an output signal, wherein ANR processing of the input signal is automatically controlled to limit a loudness level of the output signal; in response to determining that the energy of the input signal is greater than the estimate of the ambient noise by the predetermined amount, generating an output signal, wherein ANR processing of the input signal is not automatically controlled to limit the loudness level of the output signal; and driving a sound transducer of the ANR headset with the output signal.

[0024] In an example, the step of generating an output signal includes the following, wherein ANR processing of the input signal is automatically controlled to limit the loudness level of the output signal: generating an output signal, wherein ANR processing of the input signal is automatically controlled to limit the loudness level of the output signal to a level that is below or substantially equivalent to a predefined target loudness level of the output signal.

[0025] In the example, the predefined target loudness level is the sound pressure level at the user's ear of the ANR headset.

[0026] Details of one or more specific implementations are discussed in the accompanying drawings and the following description. Other features, objects, and advantages will be apparent from the specification, the drawings, and the claims. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 An example of an in-ear active noise reduction (ANR) headset is shown.

[0028] Figure 2 is a block diagram of an example configuration of an ANR device.

[0029] Figure 3A is a block diagram of an example implementation of an ANR device, in which a variable direct path is provided in parallel with the ANR path in the feedforward signal flow path.

[0030] Figure 3B is a block diagram of an example implementation of a binaural ANR system, in which the variable gain of the direct path provided in parallel with the ANR path for each ear is controlled by a coprocessor based on the estimated noise levels at both ears.

[0031] Figure 3C is a block diagram of an example implementation of an ANR device, in which multiple variable direct paths are provided in parallel with the ANR path in the feedforward signal flow path.

[0032] Figure 4A is a block diagram of an example implementation of a pulse detector.

[0033] Figure 4B is a block diagram of an example implementation of a pulse detector.

[0034] Figure 5A is a graph of the acoustic signal of a cough and the generated pulse detection markers.

[0035] Figure 5B is a graph of the detection signal and the ambient noise signal generated by the acoustic signal of a cough.

[0036] Figure 5C is a graph of the signal-to-noise ratio of the detection signal and the ambient noise signal generated by the acoustic signal of a cough and the pulse detection threshold.

[0037] Figure 6A is a graph of the acoustic signal of the box being closed and the generated pulse detection markers.

[0038] Figure 6B is a graph of the detection signal and the ambient noise signal generated by the box closing signal of a cough.

[0039] Figure 6C It is a graph of the signal-to-noise ratio and pulse detection threshold of the detection signal and the ambient noise signal generated by the acoustic signal closed by the box.

[0040] Figure 7A It is a graph of the acoustic signal of pink noise and the generated pulse detection markers.

[0041] Figure 7B It is a graph of the detection signal and the ambient noise signal generated by the acoustic signal of pink noise.

[0042] Figure 7C It is a graph of the signal-to-noise ratio and pulse detection threshold of the detection signal and the ambient noise signal generated by the acoustic signal of pink noise.

[0043] Figure 8 It is a flowchart of an example process for suppressing a noise cancellation signal output from one or more noise cancellation filters in response to an impulsive acoustic input to a microphone (such as a feedforward microphone). Detailed Description

[0044] The present disclosure relates to the use of active noise cancellation (ANR) in acoustic devices while allowing a user to perceive ambient sounds up to a threshold amount and suppressing the ANR response to fast transient sounds (also referred to herein as "pulses").

[0045] In some examples, the techniques described herein allow for the implementation of an ANR signal flow path in parallel with a variable transparency signal flow path or a through signal flow path, where the gain of the through signal path can be controlled or adjusted based on threshold conditions of the ambient noise. For example, a device implementing the technique can be configured to pass ambient sounds up to a threshold level (possibly with some parallel ANR processing), but enable or gradually increase ANR processing when the amplitude of the ambient sound exceeds the threshold. In some cases, this can improve the overall user experience by, for example, helping the user avoid excessive acoustic isolation in low-noise environments while still providing ANR functionality when the noise exceeds the threshold.

[0046] In addition, the techniques described herein can suppress the ANR response to rapid transient sounds. Noises such as applause, silverware clinking, the click of closing a box lid, coughing, closing a door can all be characterized as pulses of these sorts. Any ANR response to a pulse will almost necessarily be slower than the pulse that triggered the ANR response, meaning that the user will hear the pulse and then hear a delayed, brief noise reduction, which can be noticeable and distracting. To avoid such behavior, the ANR response to a pulse can be suppressed by comparing an input signal (e.g., from a feedforward microphone) with an estimate of the ambient noise to determine whether a rapid increase in signal energy has occurred. If such an increase has occurred, it can be determined that the ANR processing can be suppressed (e.g., temporarily frozen) so that an unwanted ANR response to the pulse does not occur.

[0047] As background, ANR devices such as active noise reduction (ANR) headphones are used to provide a potentially immersive listening experience by reducing the effects of ambient noise and sounds. However, by blocking the effects of ambient noise, ANR devices can create an acoustic isolation from the environment, which may be undesirable under some conditions. For example, a user waiting at an airport terminal may want to perceive flight announcements while using ANR headphones. Also, when using ANR headphones to cancel the noise of an airplane flight, the user may want to be able to communicate with the flight attendant without having to remove the headphones.

[0048] In addition, some headphones provide a feature commonly referred to as "pass-through" or "monitoring", where an external microphone is used to detect external sounds that the user may want to hear. For example, the external microphone can allow signals in the corresponding frequency band to be delivered through the headphones when it detects sounds in the voice frequency band of interest or some other frequency band. Some other headphones allow multi-mode operation, where in the "transparency" mode, the ANR functionality can be turned off or at least reduced in at least one frequency range to allow a relatively wideband of ambient sounds to reach the user. However, in some cases, the user may want to perceive ambient sounds up to a threshold and want the ANR processing to occur only when the ambient sound exceeds the threshold. In addition, the user may want to have a certain degree of control over the amount of ambient sound passing through the ANR device.

[0049] An active noise reduction (ANR) device can include a configurable digital signal processor (DSP) that can be used to implement various signal flow topologies and filter configurations. Examples of such DSPs are described in U.S. Patent No. 8,073,150 and U.S. Patent No. 8,073,151, which are hereby incorporated by reference in their entirety. U.S. Patent No. 9,082,388 (likewise incorporated by reference in its entirety herein) describes Figure 1Acoustic specific implementation of the in-ear active noise reduction (ANR) headset shown. The headset 100 includes a feedforward microphone 102, a feedback microphone 104, an output transducer 106 (which may also be referred to as an electroacoustic transducer or an acoustic transducer), and a noise reduction circuit (not shown), which is coupled to the two microphones and the output transducer to provide an anti-noise signal to the output transducer based on the signals detected at the two microphones. An additional input ([ Figure 1 not shown in) provides additional audio signals, such as music or communication signals, for playback on the output transducer 106 independently of the noise reduction signal. The additional input can be a wired or wireless (e.g., Bluetooth) connection to an audio source.

[0050] The term "headset", which is used interchangeably with the term "headphone" herein, includes various types of personal acoustic devices, such as in-ear, around-ear, or over-ear headphones, earphones, and hearing aids. A headset or headphone may include earbuds or earcups for each ear. The earbuds or earcups may be physically tied together (e.g., by a cord, a headband bridge, or a headband or a postauricular retention structure). In some specific implementations, the earbuds or earcups of the headset may be connected to each other via a wireless link.

[0051] Various signal flow topologies can be implemented in the ANR device to achieve functions such as audio equalization, feedback noise cancellation, feedforward noise cancellation, etc. For example, as Figure 2 shown in the example block diagram of the ANR device 200 in, the signal flow topology may include a feedforward signal flow path 110, which drives the output transducer 106 (using, for example, a feedforward compensator 112) to generate an anti-noise signal, thereby reducing the impact of the noise signal picked up by the feedforward microphone 102. Again, the signal flow topology may include a feedback signal flow path 114, which drives the output transducer 106 to generate an anti-noise signal (using, for example, a feedback compensator 116) to reduce the impact of the noise signal picked up by the feedback microphone 104. The signal flow topology may also include an audio path 118, which includes a circuit (e.g., an equalizer 120) for processing the input audio signal 108 (such as music or communication signals) for playback on the output transducer 106. In some specific implementations, the feedforward compensator 112 may include an ANR signal flow path arranged in parallel with a direct path. An example of such a configuration is described in U.S. Patent No. 10,096,313, the entire content of which is incorporated herein by reference.

[0052] In some specific implementations, the output of the output transducer 106 can be adjusted according to the desired final volume or loudness at the ear, such that the total attenuation provided by the ANR device (e.g., obtained by controlling one or both of the ANR and direct signal paths) increases and decreases as the ambient noise level increases and decreases, respectively. For example, when the ambient sound level does not meet a threshold condition (e.g., is below a threshold level), the ambient sound can be allowed to pass through to the ear with little or no attenuation. On the other hand, when the ambient sound level meets the threshold condition (e.g., breaches the threshold level), the ambient sound can be attenuated, perhaps progressively (i.e., with more attenuation as the environment gets louder).

[0053] Figure 3A FIG. 4 is a block diagram of an example specific implementation of the ANR device 300, where a variable direct path is provided in parallel with the ANR path in the feedforward signal flow path to provide the variable attenuation described above. Specifically, the device 300 includes an ANR filter 305 (also denoted as KANR) provided in parallel with a combination of a direct filter 310 (also denoted as KAW) and a detector filter 315 (also denoted as and referred to as a side-chain filter Kd). The detector filter 315 can be used to monitor the signal captured using the FF microphone 102 and control the input to the direct filter (e.g., using a variable gain amplifier (VGA) or compressor 320). In some specific implementations, the input to the detector filter 315 can be preprocessed, e.g., to make the detector filter 315 more sensitive to certain types of signals. For example, the side-chain filter can be configured to make the detector filter more sensitive to changes in the perceived weighted speech band noise level. The output of the detector filter 315 can be used to adjust the VGA 320, which applies a gain to the input signal provided to the direct filter 310.

[0054] In some specific embodiments, the detector filter 315 may include a frequency weighting filter (e.g., an A-weighting filter and / or a filter representing the head-related transfer function (HRTF)). The detector filter 315 may also include a level generator that converts the output of the frequency weighting filter into a signal level and then compares the signal level with a threshold level (e.g., a user-defined or pre-determined level). The detector filter 315 may also include a signal generator configured to generate a control signal for controlling the gain of the VGA 320. In some specific embodiments, the signal generator may be configured to generate the control signal dynamically according to the target attack and decay rates. The "attack rate" is defined as the rate at which the attenuation increases. In some specific embodiments, the target attack rate is less than 100 dB per second (total insertion gain), such as approximately 10 dB / second. The "decay rate" or "release rate" is defined as the rate at which the attenuation decreases. In some specific embodiments, the decay rate is two times or more faster than the attack rate. In some specific embodiments, the combination of a low threshold (e.g., <80 dBA of the insertion gain) and a low attack rate (e.g., <100 dB / second) can be used for a comfortable user experience in various scenarios of daily life.

[0055] In some specific embodiments, the detector filter 315 may be configured to control the VGA or the compressor 320 according to threshold conditions. The threshold conditions may be preset or set according to user input. In some specific embodiments, if the detector filter 315 determines that the ambient noise level is below a specific threshold, the output of the detector filter 315 controls the compressor or the VGA 320 such that the gain of the direct signal flow path is substantially equal to one. This in turn allows the user to hear the ambient sound with substantially little or no attenuation. In some specific embodiments, if the detector determines that the ambient noise level is at or above the threshold, the output of the filter 315 may be configured to control the compressor or the VGA 320 such that the total gain of the direct signal path is less than one, and the output of the ANR filter 305 provides attenuation of the noise at the ear. This allows the user to perceive the ambient noise and sound when the noise is below the threshold, and to utilize the ANR functionality of the headset when the noise breaches the threshold - for example, to prevent loud sounds from sources such as vehicles, sirens, or machines from becoming uncomfortably loud.

[0056] While Figure 3AThe example of only shows the VGA disposed in the direct signal path, but other variations are possible. For example, the VGA can be disposed in the ANR path, as a supplement or alternative to the VGA 320 disposed in the direct signal path. In some specific embodiments, the VGA disposed in the signal path (e.g., the ANR path or the direct path) can be controlled to adjust the weight associated with the corresponding path. For example, the VGA gain can be set to be substantially equal to zero to make the weight associated with the corresponding path be substantially equal to zero. In some specific embodiments, one or more additional parameters associated with the VGA (or generally the corresponding path) can be adjusted to control one or more characteristics of the corresponding path. For example, the response rate of the VGA (which can also be referred to as the compressor attack and release times corresponding to whether the compression is increasing or decreasing step by step, respectively) can be adjusted to provide a fast response or a relatively gradual response to the changing noise level. In some specific embodiments, this can specify how fast the ANR device adjusts the gain when the noise level meets the threshold condition, and / or how fast the ANR device reduces the gain or restores it to a predetermined level (e.g., one) when the noise level no longer meets the threshold condition. In some specific embodiments, the response rate can be adjusted such that the ANR processing smoothly responds to the increase in the noise level based on the target attack rate. In some specific embodiments, the target attack rate can be less than 100 dB / second.

[0057] In some specific embodiments, the outputs of the ANR path and the direct path (e.g., in a weighted combination) are combined to generate a feedforward signal 325 that at least partially drives the acoustic transducer 106. In some specific embodiments, the feedforward signal 325 can be combined with the feedback signal 330 and / or one or more other signals 335. The signal 335 can include, for example, a media signal derived from the audio input 108 or a signal from one or more other microphones or audio sources.

[0058] In some specific embodiments, the gain control of the VGA or the compressor 320 in each of the two separate earbuds or earcups can be coordinated, for example, to avoid having substantially unequal noise reduction in the two earbuds / earcups of the headset. Figure 3B is a block diagram of an example specific implementation of such a binaural ANR system 350, where the variable gain of the direct path arranged in parallel with the ANR path for each ear is controlled based on the estimated noise levels at both ears. Specifically, Figure 3BThe specific implementation shown includes a coprocessor 360 that receives inputs from noise estimator modules 355 disposed in each of two earbuds or earcups 352a and 352b (generally 352), and coordinates the gain control of corresponding VGAs or compressors 320 in the two earbuds or earcups 352. In some specific implementations, the coprocessor 360 is disposed in one of the earbuds or earcups 352. In some specific implementations, the coprocessor 360 may be disposed in a device external to the headset, such as a device that is the source of the acoustic media being played through the headset. The coprocessor may include one or more processing devices configured to analyze the inputs received from the noise estimator 355 and generate gain control signals for the VGA 320.

[0059] In some specific implementations, the noise estimator 355 includes one or more digital filters configured to generate a signal providing an estimate of the noise at the location of the corresponding earbud or earcup 352. For example, the noise estimator 355 may include a front-end weighting filter that emphasizes the part of the spectrum that is most indicative of the loudness at which sound is perceived. In some specific implementations, the response of the front-end weighting filter is approximately divided by the A-weighting of the head-related transfer function (HRTF) (or another function representing the effect of the presence / orientation of the user's head) to refer to the noise signal measured at the earbud microphone of the headset as a diffuse field. Other front-end weighting filters (such as B-weighting or C-weighting) are possible, or a more complex loudness model may be used. In some specific implementations, the front-end weighting filter may be used to compensate for hardware effects (e.g., microphone sensitivity). In some specific implementations, the front-end weighting filter may include a plurality of cascaded filters, each of the plurality of cascaded filters addressing and / or compensating for individual effects (e.g., the effect of the presence / orientation of the head, the effect of the hardware, and / or A-weighting). The output of the weighting filter may be an AC signal representing the relative loudness perceived at the corresponding ear. Such output may then be post-processed (e.g., by rectifying and then low-pass filtering) before being provided to the coprocessor 360 as an estimate of the noise level at the corresponding ear.

[0060] In some specific implementations, Figure 3A and Figure 3B The depicted system may be implemented as part of a multi-band system having two or more parallel paths, each parallel path having its own VGA 320 and through-filter KAW 310, all of which are disposed in parallel with the KANR 305. Figure 3C An example of such a device is depicted, which shows Figure 3A a multi-band version of the device. Specifically, Figure 3CIt is a block diagram of an exemplary embodiment of an ANR device 375, in which multiple variable direct paths are arranged in parallel with the ANR path in the feedforward signal flow path. Each path includes a corresponding direct filter (one of 310a,...... and 310n, generally 310), a corresponding detector filter (one of 315a,...... and 315n, generally 315), and a corresponding VGA (one of 320a,...... and 320n, generally 320). Each direct filter 310 passes through different parts of the desired direct spectrum (filtered, for example, using one of the corresponding bandpass filters - 380a,...... and 380n), such that when all VGAs 320 have unit gain, an overall desired "perceived" response is achieved. In some embodiments, the various parameters of the different parallel paths can be configured individually. For example, a particular parallel path can be configured to have its own attack rate and release rate, compression ratio, and / or threshold suitable for the corresponding frequency band. In some embodiments, one or more parameters (e.g., threshold and compression ratio) can be common across multiple parallel paths, while the corresponding attack rate and release rate can be different. This can allow for frequency-specific tuning of the response of the ANR device. For example, the device can be configured to have a fast response to high-frequency noise spikes but a relatively slow response to low-frequency noise. In some embodiments, the parameters of the different paths can be made user-adjustable.

[0061] In some embodiments, the components of the feedforward signal path 110 can be adjusted in various ways to generate the feedforward signal 325. US11087776 shows such methods of adjusting these and illustrates graphs of some exemplary variations of ANR processing performed in the feedforward signal path 110 based on different threshold conditions, and the entire patent application is incorporated herein by reference.

[0062] In some examples, the response of the ANR path may be briefly suppressed to avoid responding to detected pulses (i.e., fast transient signals characterized by a sharp increase and corresponding sharp decrease in noise, typically within 1 ms to 2 ms). To suppress the ANR response to a pulse, it is necessary to first distinguish the pulse from ambient noise (i.e., noise that should ideally be filtered out, such as the low-frequency hum of aircraft noise or the sound of a passing motorcycle). To achieve this, the detector filter 315 may also be configured to compare the energy in each sample (i.e., the sample to be tested) with the ambient noise estimated energy. The ambient noise may be estimated by characterizing the energy in adjacent samples. If the energy in the sample to be tested is greater than the energy in adjacent samples by some predetermined amount, a large spike in energy indicative of a pulse may be inferred. Of course, pulses that are not greater than the ambient noise may occur, but in these cases, the pulse will likely not be perceptible through the ambient noise and should not interfere with the ANR gain. (Although described in connection with the detector filter 315, it should be understood that pulse detection may be performed at any suitable location within the topology.)

[0063] The energy in adjacent samples may be estimated in a variety of ways. In one example, a buffer of samples may be stored and the values of each averaged. The buffer of samples may include samples before the sample to be tested, samples after the sample to be tested, or both. In either case, it is generally useful (although not required) to exclude the sample to be tested itself from the average, as it will tend to skew the average relative to the measured value of the sample to be tested. In an alternative example, an exponentially weighted moving average may be used to average the weighted average of the current sample and previous samples, rather than using a buffer of samples.

[0064] In an alternative example, the input signal may be transformed into the frequency domain (e.g., by DFT or other suitable frequency transformation) rather than operating in the time domain. The ambient noise estimate may be determined based on the average power in the resulting frequencies or in a subset of the frequency intervals of interest. This average power may be updated for each consecutive frame of the frequency interval (i.e., for each new sample) (e.g., by an exponentially weighted moving average) or may be calculated independently for each frame. Any specified interval or set of intervals above the calculated ambient noise may be flagged as a pulse. However, these methods are relatively memory-intensive and computationally expensive.

[0065] In an alternative example, as Figure 4AAs shown, two low-pass filters can be employed in parallel paths. The low-pass filters of the two paths can smooth-filter the input signal at different rates. For example, the low-pass filter 402 can apply a greater smoothing filter to the input signal than the low-pass filter 404, thereby generating a signal that will represent an estimated value of the ambient noise. (In this example, the output of this low-pass filter 402 is thus referred to as the ambient noise signal 406.) More specifically, in some examples, the output of the low-pass filter 404 can be customized, similar to an exponentially weighted moving average, to provide an approximate average of the samples in a sliding window. Additionally, in various examples, the output of the low-pass filter 402 can be scaled and / or other processing can be performed on it to customize it to appropriately represent the ambient noise. In these examples, the ambient noise signal 406 is the cumulative output of the process performed to estimate the ambient noise, rather than just the output of the low-pass filter 402. In contrast, the output of the low-pass filter 404 can apply a relatively fast smoothing filter, which operates as an envelope detector to characterize the peaks of the input signal. This filter is used to spread the pulses such that the estimated values of the peaks are more easily captured and compared with the ambient noise estimate. The output of this filter 404 is referred to as the detection signal 408. The detection signal 408 can also be the result of additional processing that characterizes the energy of the sample to be tested in a manner that aids in pulse detection. Additionally, in certain examples, the low-pass filter 404 can be omitted and the input signal can be directly relied upon to be compared with the ambient noise signal 406. For the purposes of this disclosure, in examples where the low-pass filter 404 is omitted, the input signal becomes the detection signal 408.

[0066] Generally speaking, any suitable low-pass filter can be used for the low-pass filter 402 and the low-pass filter 404. Additionally, to achieve different smoothing filter characteristics for the low-pass filters 402 and 404, different cut-off frequencies can be selected for each filter. For example, the cut-off frequency of the low-pass filter 402 can be set to 5 Hz, while the cut-off frequency of the low-pass filter 404 can be set to 100 Hz, although other suitable cut-off frequencies can also be used. In various examples, alternatively, the input signal can be filtered with a FIR Hilbert transform or with a whitening filter, which helps remove any spectral shape of the ambient noise to produce a more robust pulse detection (although these examples may require more processing power than is typically available).

[0067] Since the environmental noise signal 406 represents an estimate of the environmental noise, the presence of a pulse can be detected by comparing the detection signal 408 with the environmental noise signal 406. The comparison of the detection signal 408 with the environmental noise signal 406 can be done in one of a variety of ways. In one example, the difference between the detection signal 408 and the environmental noise signal 406 can be found by a difference module 410, and the output of this difference module is input to a comparator 412 for comparing the difference between these signals with a threshold. If the difference between the signals is greater than the threshold, the input signal can be marked as possibly containing a pulse or at least containing the start of new continuous noise. In an alternative example, the ratio of the detection signal 408 and the environmental noise signal 406 can be found and compared with a threshold to determine whether the ratio of the two signals indicates a pulse, rather than finding the difference between the two signals. (This method is similar to comparing the signal-to-noise ratio of two signals with a threshold.) Other suitable methods of comparing the detection signal 408 with the environmental noise signal 406 are envisioned herein, which give some indication of how much larger the detection signal 408 is than the environmental noise signal 406.

[0068] In Figure 4B the example shown, to better represent the environmental noise, a delay 414 can be used to delay the output of the low-pass filter 402 by a predetermined amount (e.g., 2 ms) to prevent the sample to be tested from affecting the environmental noise signal 406 with which it is compared. In other words, by applying the delay, the environmental noise signal represents the environmental noise that existed prior to the current sample and thus better represents the environmental noise with which the signal should be compared. Generally, such a delay is useful for capturing all energy pulses except for very fast energy pulses, which can be captured without a delay.

[0069] Furthermore, as Figure 4A and Figure 4B shown, the input signals applied to the low-pass filter 402 and the low-pass filter 404 can be the absolute value (e.g., the rectified version) of the output of the front-end microphone 102. This is to prevent natural oscillating audio signals from deviating from the average of the environmental noise and to ensure that the detection signal 408 and the environmental noise signal 406 have the same sign when being compared. Additionally, it should be understood that additional processing (e.g., a high-pass filter) can be performed on the input signals before they are received at the low-pass filters 402 and 404 such that pulses can be detected more reliably, or additional processing can be performed for any other suitable reason.

[0070] In response to the output of comparator 412 indicating that a pulse may be present, the ANR response generated by the feedforward microphone output samples containing the detected pulse can be suppressed. However, if the comparator 412 output does not indicate the possible presence of a pulse, the output of the ANR is not suppressed and instead follows the system's parameters to apply ANR based on threshold conditions or some other metric, as described above. If the tagged sample is not a pulse but the start of a new continuous noise (e.g., an approaching motorcycle), the initial sample containing the new continuous noise will be higher than the ambient noise and will therefore initially be tagged as a pulse. However, as the ambient noise continues, the ambient noise signal will quickly increase to the level of the detection signal, meaning that the comparison result of the two signals will only briefly exceed the threshold condition. The time constant of the ANR filter is typically such that the delay will likely not be noticed by the user.

[0071] To further improve the performance of pulse detection, each sample exceeding the threshold can be zeroed or otherwise adjusted so that the detected pulse does not affect the ambient noise signal 406. In other words, during the delay implemented by delay unit 414, the detected energy of the pulse can be removed so that the detected pulse does not affect the background noise measurement of future samples.

[0072] To illustrate the operation of pulse detection, FIGS. 5 through 7 depict various recorded audio signals, the output detection signal 408 and the ambient noise signal 406, and the calculated signal-to-noise ratio between the two. Figure 5A Depicts an input audio signal of a person coughing at fairly regular intervals (approximately one second apart). Figure 5B Depicts the detection signal 408 and the ambient noise signal 406, which have been delayed by 2 ms. As shown, since the ambient noise signal 406 employs a relatively large amount of smoothing filtering, the spikes of the input signal are not captured. Additionally, in terms of the peaks of the input signal being captured, the detection signal and the ambient noise signal are delayed by 2 ms so that the difference between the captured peaks in the detection signal 408 is compared to the ambient noise present before the start of the cough. Thus, as Figure 5C shown, the peak at the start of the cough is sufficient to produce a large difference between the detection signal 408 and the ambient noise signal 406, thus triggering a detection flag ( Figure 5A shown) and suppressing the ANR response. It should be noted that in this example, the detection flag is held for up to a predetermined period after the SNR exceeds the 15 dB threshold to ensure the full duration of the ANR response to the pulse.

[0073] In a similar manner, Figures 6A to 6C depicts the output of the feedforward microphone 102 for the periodic closing of a box. Figure 6Bshows the detection signal 408 and the ambient noise signal 406 generated by this signal, similar to Figure 5A the cough signal of Figure 6C which results in an initial peak in the detection signal sufficient to generate an SNR in

[0074] that exceeds a predetermined threshold and sets a flag to suppress the ANR response. Figure 7A In contrast, Figure 7B depicts an audio signal generated by a telephone that starts playing pink noise at about the 0.75 - second mark and then is waved back and forth near the feed - forward microphone 102. As shown, at the initial start of the pink - noise audio signal, a spike in the detection signal 408 relative to the ambient noise signal 406 ( Figure 7C ) records an SNR that triggers the suppression of the ANR response. However, afterwards, the changes in the audio signal due to waving the telephone back and forth do not create a large enough difference between the detection signal 408 and the ambient noise signal 406 to exceed the SNR threshold. This shows that while the initial start of a new continuous ambient noise will trigger and briefly delay the ANR response, the ambient noise signal 406 quickly adapts, allowing the ANR response to reduce the ambient noise as desired.

[0075] ANR suppression can occur in any of a suitable manner and will depend in part on the particular implementation of the ANR / through - path system. In one example, an interrupt signal can be generated that temporarily freezes the adjustment of the VGA 320 such that the ANR response remains constant until the time period has passed during which the adjustment generated by the pulse has occurred. Alternatively, the response time of the VGA 320 associated with the signal path (e.g., as described in reference Figure 3A ) can be adjusted. The response rate of the VGA 320 can be changed such that the ANR is adjusted relatively gradually in response to the sample to be tested, such that the change in the ANR is less or not noticeable to the user. It should be understood that for the purposes of this disclosure, suppressing the ANR response means reducing the ANR response relative to how the ANR response would occur during normal operation (i.e., without any intervention from pulse detection) in response to a pulse. Thus, both keeping the ANR response constant once a pulse is detected or slowing down the response to the pulse are considered to be "suppressing" the ANR response.

[0076] Depending on the topology of the ANR system, suppressing the ANR response may require adjusting or keeping constant the VGA at the input or output of the ANR filter. Additionally, the ANR filter itself can be adjusted, such as adjusting its adaptation rate so that it does not adapt or adapts very slowly to the incoming pulse. Other suitable methods of suppressing the ANR response (such as filtering the input samples from the ANR filter input) are conceivable and within the scope of this disclosure.

[0077] In addition, in an alternative example, rather than suppressing the entire ANR response, the ANR response can be adjusted to mitigate the effects of overloading transducer 106 in response to a large input signal. A large input signal can cause microphone clipping, which introduces noise transients into the voltage signal applied to the transducer. A large input also tends to result in a large ANR response that overloads transducer 106. The overloading of transducer 106 can be mitigated by reducing the ANR filter gain in certain portions of the frequency range (e.g., very high or very low frequencies). Other measures for mitigating transducer overloading are conceivable and, like the example of suppressing the ANR output, depend in part on the topology of the ANR / through system.

[0078] The pulse detection described in this disclosure can further be used to control the operation of a device such as headset 100. For example, the digital signal processor can also be programmed to monitor a pulse detection flag corresponding to a set of pulses for a preset user input. As an example, the digital signal processor can be programmed to treat two separate pulses (spaced approximately half a second apart) as a user command to pause the audio track, skip to the next track, etc. This is provided only as an example of the kind of pulses that can be considered a user command. In general, it is beneficial to select pulses that are easy for the user to generate, such as by clicking the user's tongue, and that are unlikely to occur outside of an intentional command.

[0079] Figure 8 FIG. 800 is a flow diagram of an example process 800 for suppressing a noise cancellation signal output from one or more noise cancellation filters in response to a pulsed acoustic input to a microphone (such as a feedforward microphone).

[0080] At least a portion of process 800 can be implemented using one or more processing devices (such as the DSPs described in U.S. Patent Nos. 8,073,150 and 8,073,151, which are incorporated herein by reference in their entirety). In some embodiments, process 800 can be implemented in a device having a signal path that is substantially similar to those depicted in FIGS. 3-4.

[0081] At step 802, an input signal is received from one or more microphones based on ambient sound captured. The microphone can be, for example, a feedforward microphone such as feedforward microphone 102. At step 804, the input signal is processed by one or more noise cancellation filters (e.g., ANR filters) to generate a noise cancellation signal configured to reduce the effects of the input signal.

[0082] At step 806, the input signal is compared with an estimate of the ambient noise to determine whether the energy of the input signal is greater than the estimate of the ambient noise. The estimate of the ambient noise can be made in a variety of ways. For example, a buffer of samples can be stored and the values of each averaged, or an exponentially weighted moving average can be used to average the current sample with a weighted average of previous examples. Alternatively, the input signal can be transformed into the frequency domain and the power of a frequency bin or subset of frequency bins averaged to determine the ambient noise estimate. In yet another example, a low-pass filter such as that shown in Figure 4A and Figure 4B can be used to smooth the input signal and form an estimate of the ambient noise.

[0083] The energy of the input signal (which itself can be output from a low-pass filter, such as that shown in Figure 4A and Figure 4B ) can be compared with the estimate of the ambient energy by finding, for example, the difference or ratio between the energy of the input signal and the estimated ambient energy.

[0084] At step 808, when it is determined that the energy of the input signal is greater than a predetermined amount greater than the estimate of the ambient noise, the noise reduction signal is suppressed. The comparison result in step 806 can be compared with a threshold to determine whether it indicates a pulse (or at least the start of continuous noise). If the comparison result exceeds the threshold, action can be taken to suppress the noise reduction signal.

[0085] ANR suppression can occur in any of a suitable number of ways and will depend in part on the particular implementation of the ANR / through system. In the examples of FIGS. 3 to 4, an interrupt signal can be generated that temporarily freezes the adjustment of the variable gain amplifier such that the ANR response remains constant until the time period during which the adjustment due to the pulse has occurred has passed. Alternatively, the response time of the VGA 320 associated with the signal path (e.g., as described with reference to Figure 3A ) can be adjusted. The response rate of the VGA 320 can be varied such that the ANR is adjusted relatively gradually in response to the sample being tested such that the change in the ANR is less or not noticeable to the user.

[0086] In addition, depending on the topology of the ANR system, suppressing the ANR response may also require adjusting or holding constant the VGA at the input or output of the ANR filter. In addition, the ANR filter itself can be adjusted, such as adjusting its adaptation rate such that it does not adapt to the incoming pulse. Other suitable methods of suppressing the ANR response (such as filtering the input samples from the ANR filter input) are conceivable and within the scope of the present disclosure.

[0087] At step 810, an output signal is generated for the acoustic transducer, the output signal at least partially including a noise reduction signal such that the acoustic transducer generates an acoustic output in accordance with the output signal.

[0088] The functionality described herein, or portions thereof, and various modifications thereof (hereinafter referred to as “functionality”) can be implemented at least in part via a computer program product (e.g., a computer program tangibly embodied in an information carrier, such as one or more non-transitory machine-readable media or storage devices) for execution by, or to control the operation of, one or more data processing devices (e.g., programmable processors, computers, multiple computers, and / or programmable logic components).

[0089] The computer program can be written in any form of programming language, including compiled or interpreted languages, and the computer program can be deployed in any form, including as a stand-alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. The computer program can be deployed on one computer or distributed across one site or multiple sites and executed on multiple computers interconnected by a network.

[0090] Actions associated with implementing all or part of the functionality can be performed by one or more programmable processors executing one or more computer programs to perform the functions of the calibration process. All or part of the functionality can be implemented as special purpose logic circuitry, such as FPGAs and / or ASICs (application specific integrated circuits).

[0091] Processors suitable for the execution of a computer program include, by way of example, both general and special purpose microprocessors, and any one or more processors of any type of digital computer. In general, a processor will receive instructions and data from a read only memory or a random access memory or both. Components of a computer include a processor for executing instructions and one or more memory devices for storing instructions and data.

[0092] Elements of the different embodiments described herein can be combined to form other embodiments not specifically recited above. Some elements can be removed from the structures described herein without adversely affecting their operation. In addition, various separate elements can be combined into one or more separate elements to perform the functionality described herein.

Claims

1. An apparatus, the apparatus comprising: Noise-canceling headphones, the noise-canceling headphones comprising one or more microphones and a sound transducer, the one or more microphones being configured to generate an input signal based on captured ambient sound; and A controller, the controller comprising one or more processing devices, the controller being configured to: Process the input signal through one or more noise-canceling filters to generate a noise-canceling signal, wherein the noise-canceling signal is configured to reduce the impact of the input signal; Compare the input signal with an estimate of ambient noise to determine whether the energy of the input signal is greater than the estimate of ambient noise, wherein if the energy of the input signal is greater than the estimate of ambient noise by a predetermined amount, the change in the noise-canceling signal is suppressed; and Generate an output signal, the output signal at least partially comprising the noise-canceling signal, wherein the sound transducer is configured to produce a sound output according to the output signal.

2. The apparatus according to claim 1, wherein the output signal is a weighted combination of the noise-canceling signal and a through signal.

3. The apparatus according to claim 1, wherein comparing the input signal with an estimate of ambient noise comprises: Comparing the energy of the input signal with an ambient noise signal generated by a first low-pass filter, wherein the first low-pass filter is configured such that the ambient noise signal is an estimate of the ambient noise present in the captured ambient sound.

4. The apparatus according to claim 1, wherein the ambient noise signal is delayed in time with respect to the input signal.

5. The apparatus according to claim 3, wherein the energy of the input signal is determined by the output of a second low-pass filter, wherein the second low-pass filter has a greater impact on the smoothing filtering of the input signal than the first low-pass filter.

6. The apparatus according to claim 5, wherein comparing the energy of the input signal further comprises: Determining whether a difference between the output of the second low-pass filter and the ambient noise signal satisfies a threshold condition.

7. The apparatus according to claim 5, wherein comparing the energy of the input signal further comprises: Determining whether a ratio between the output of the second low-pass filter and the ambient noise signal satisfies a threshold condition.

8. The apparatus according to claim 1, wherein suppressing the noise-canceling signal comprises: Temporarily stopping to increase the amplitude of the noise-canceling signal.

9. The apparatus according to claim 8, wherein temporarily stopping to increase the amplitude of the noise-canceling signal comprises: Temporarily stopping to adjust a variable gain filter in a through-processing chain for generating a through signal, wherein the output signal is a weighted combination of the noise-canceling signal and the through signal.

10. The apparatus according to claim 1, wherein suppressing the noise-canceling signal comprises: Adjusting the rate at which the noise-canceling signal is adjusted in response to the input signal.

11. One or more non-transitory machine-readable storage devices having computer-readable instructions encoded thereon for causing one or more processing devices to perform a method, the method comprising the steps of: Receiving an input signal from one or more microphones, the input signal being based on captured ambient sound; Processing the input signal through one or more noise reduction filters to generate a noise reduction signal, wherein the noise reduction signal is configured to reduce the impact of the input signal; Comparing the input signal with an estimate of ambient noise to determine whether the energy of the input signal is greater than the estimate of ambient noise, wherein if the energy of the input signal is greater than the estimate of ambient noise by a predetermined amount, the change in the noise reduction signal is suppressed; and Generating an output signal for a sound transducer, the output signal at least partially comprising the noise reduction signal, such that the sound transducer produces an acoustic output in accordance with the output signal.

12. The one or more non-transitory machine-readable storage devices of claim 11, wherein the output signal is a weighted combination of the noise reduction signal and a direct-through signal.

13. The one or more non-transitory machine-readable storage devices of claim 11, wherein comparing the input signal with an estimate of ambient noise comprises: Comparing the energy of the input signal with an ambient noise signal generated by a first low-pass filter, wherein the first low-pass filter is configured such that the ambient noise signal is an estimate of the ambient noise present in the captured ambient sound.

14. The one or more non-transitory machine-readable storage devices of claim 11, wherein the ambient noise signal is delayed in time relative to the input signal.

15. The one or more non-transitory machine-readable storage devices of claim 13, wherein the energy of the input signal is determined by the output of a second low-pass filter, wherein the second low-pass filter has a greater impact on smoothing the input signal than the first low-pass filter.

16. The one or more non-transitory machine-readable storage devices of claim 15, wherein comparing the energy of the input signal further comprises: Determining whether a difference between the output of the second low-pass filter and the ambient noise signal satisfies a threshold condition.

17. The one or more non-transitory machine-readable storage devices of claim 15, wherein comparing the energy of the input signal further comprises: Determining whether a ratio between the output of the second low-pass filter and the ambient noise signal satisfies a threshold condition.

18. The one or more non-transitory machine-readable storage devices of claim 15, wherein suppressing the noise reduction signal comprises: Temporarily stopping to increase the amplitude of the noise reduction signal.

19. The one or more non-transitory machine-readable storage devices of claim 18, wherein temporarily stopping to increase the amplitude of the noise reduction signal comprises: Temporarily stop to adjust a variable gain filter in a direct path processing chain for generating a direct path signal, wherein the output signal is a weighted combination of the noise reduction signal and the direct path signal.

20. One or more non-transitory machine-readable storage devices according to claim 11, wherein suppressing the noise reduction signal comprises: Adjusting the rate at which the noise reduction signal is adjusted in response to the input signal.

21. A method, the method comprises: Receiving an input signal representing audio captured by a microphone of an active noise reduction (ANR) headset; Processing a portion of the input signal by one or more processing devices to determine a noise level in the input signal; Determining that the noise level satisfies a first threshold condition; Comparing the input signal with an estimate of ambient noise to determine whether the energy of the input signal is greater than the energy of the estimate of the ambient noise by a predetermined amount, In response to determining that the noise level satisfies the first threshold condition and the energy of the input signal is not greater than the energy of the estimate of the ambient noise by the predetermined amount, generating an output signal, wherein the ANR processing of the input signal is automatically controlled to limit the loudness level of the output signal; In response to determining that the energy of the input signal is greater than the energy of the estimate of the ambient noise by the predetermined amount, generating an output signal, wherein the ANR processing of the input signal is not automatically controlled to limit the loudness level of the output signal; and Using the output signal to drive a sound transducer of the ANR headset.

22. The method according to claim 21, wherein the step of generating the output signal comprises the following, wherein the ANR processing of the input signal is automatically controlled to limit the loudness level of the output signal: Generating an output signal, wherein the ANR processing of the input signal is automatically controlled to limit the loudness level of the output signal to a level that is below or substantially equivalent to a predefined target loudness level of the output signal.

23. The method according to claim 22, wherein the predefined target loudness level is a sound pressure level at the ear of a user of the ANR headset.

Images