HK1220310B - Active noise reducing headphone and method of operating active noise reducing headphone - Google Patents

Description

Active noise-reducing headphones and method of operating active noise-reducing headphones

Background Art

The present disclosure relates to providing natural hear-through in active noise reduction (ANR) headphones, reproducing audio signals simultaneously with hear-through in ANR headphones, and eliminating occlusion effects in ANR headphones.

Noise reducing headphones are used to block ambient noise from reaching the user's ears. Noise reducing headphones can be active, i.e. ANR headphones, in which electronic circuits are used to generate an anti-noise signal that destructively interferes with the ambient noise, thereby canceling it, or noise reducing headphones can be passive, in which the headphones physically block and attenuate the ambient sound. Most active headphones also provide passive noise reduction measures. Headphones used for communication or for listening to entertainment audio can include one or both of active and passive noise reduction capabilities. ANR headphones can use the same speakers for audio (which includes both communication and entertainment) and cancellation, or can have separate speakers for each.

Some headphones offer a feature commonly referred to as "talk-through" or "monitoring," in which external microphones are used to detect external sounds that the user may want to hear. Those sounds are reproduced by speakers inside the headphones. In ANR headphones with talk-through functionality, the speakers used for talk-through may be the same speakers used for noise cancellation, or may be additional speakers. The external microphones may also be used for feed-forward active noise cancellation, picking up the user's own voice for communication purposes, or the external microphones may be dedicated to providing talk-through. Typical talk-through systems apply only minimal signal processing to the external signal, and we refer to these as "direct talk-through" systems. Sometimes direct talk-through systems use a bandpass filter to limit external sounds to the voice band or some other band of interest. The direct talk-through feature may be manually triggered or may be triggered by the detection of a sound of interest, such as a voice or an alarm.

Some ANR headphones include a feature that temporarily reduces noise cancellation so that the user can hear the environment, but they do not also provide talk-through. Instead, they rely on enough sound passively passing through the headphones to make the environment audible. We call this feature passive listening.

Summary of the Invention

In general, in some aspects, active noise reducing headphones include an earcup configured to couple to a wearer's ear to define an acoustic volume comprising a volume of air within the wearer's ear canal and a volume within the earcup, a feedforward microphone acoustically coupled to an external environment and electrically coupled to a feedforward active noise cancellation signal path, a feedback microphone acoustically coupled to the acoustic volume and electrically coupled to a feedback active noise cancellation signal path, an output transducer acoustically coupled to the acoustic volume via the volume within the earcup and electrically coupled to both the feedforward and feedback active noise cancellation signal paths, and a signal processor configured to apply filters and control gains of both the feedforward and feedback active noise cancellation signal paths. The signal processor is configured to apply a first feedforward filter to the feedforward signal path and a first feedback filter to the feedback signal path in a first operating mode that provides effective cancellation of ambient sounds, and to apply a second feedforward filter to the feedforward signal path in a second operating mode that provides active hear-through of ambient sounds with ambient naturalness.

Various embodiments may include one or more of the following. The feedforward filter may cause the earphone to have an overall system response at the wearer's ear that may be smooth and piecewise linear. The difference in overall noise reduction of speech noise between the first operating mode and the second operating mode may be at least 12 dBA. The second feedforward filter may have a value K _ht selected to cause the equation to be approximately equal to a predetermined target value. The signal processor may be further configured to apply a second feedback filter, different from the first feedback filter, to the feedback signal path during the second operating mode. The combination of the feedback signal path and the ear cup may reduce ambient noise reaching the entrance of the ear canal by at least 8 dB at all frequencies between 100 Hz and 10 kHz. The feedback signal path may be operational over a frequency range extending above 500 Hz. The second feedforward filter may cause the overall system response to be smooth and piecewise linear in a frequency region extending above 3 kHz. The second feedforward filter may cause the overall system response to be smooth and piecewise linear in a frequency region extending below 300 Hz. The feedback signal path may be implemented in a digital signal processor and may have a latency of less than 250 μs. The second feedforward filter defines a non-minimum phase zero in a transfer function that characterizes the feedforward signal path.

The signal processor may be further configured to apply a third feedforward filter to the feedforward signal path during active hear-through of ambient noise having an overall response different from that provided in the second operating mode. A user input may be provided, such that the signal processor is configured to select between the first, second, or third feedforward filters based on the user input. The user input may include a volume control. The signal processor may be configured to automatically select between the second and third feedforward filters. The signal processor may be configured to select between the second and third feedforward filters based on a time-averaged measurement of the level of ambient noise. The signal processor may be configured to select between the second and third feedforward filters upon receiving a user input invoking hear-through mode. The signal processor may be configured to periodically select between the second and third feedforward filters.

The signal processor may be a first signal processor and the feedforward signal path may be the first feedforward signal path, wherein the headphone includes a second earcup configured to couple to a wearer's second ear to define a second acoustic volume comprising a volume of air within the wearer's second ear canal and a volume within the second earcup, a second feedforward microphone acoustically coupled to an external environment and electrically coupled to a second feedforward active noise cancellation signal path, a second feedback microphone acoustically coupled to the second acoustic volume and electrically coupled to a second feedback active noise cancellation signal path, a second output transducer acoustically coupled to the second acoustic volume via the volume within the second earcup and electrically coupled to both the second feedforward and second feedback active noise cancellation signal paths, and a second signal processor configured to apply filters and control gains of both the second feedforward and second feedback active noise cancellation signal paths. The second signal processor may be configured to apply a third feedforward filter to the second feedforward signal path and the first feedback filter to the second feedback signal path during a first operating mode of the first signal processor, and to apply a fourth feedforward filter to the second feedforward signal path during a second operating mode of the first signal processor. The first and second signal processors may be part of a single signal processing device. The third feedforward filter may not be identical to the first feedforward filter. Only one of the first or second signal processors may apply the respective second or fourth feedforward filter to the corresponding first or second feedforward signal path during the third operating mode.The third operating mode may be activated in response to a user input.

The first signal processor may be configured to receive a crossover signal from a second feedforward microphone, apply a fifth feedforward filter to the crossover signal, and insert the filtered crossover signal into the first feedforward signal path. The signal processor may be further configured to apply a single-channel noise reduction filter to the first feedforward signal path during a second operating mode. The signal processor may be configured to detect a high-frequency signal in the feedforward signal path, compare the amplitude of the detected high-frequency signal to a threshold indicating a positive feedback loop, and activate a compression limiter if the amplitude of the detected high-frequency signal is greater than the threshold. The signal processor may be configured to gradually reduce the amount of compression applied by the limiter when the amplitude of the detected high-frequency signal is no longer above the threshold, and increase the amount of compression to a minimum level at which the amplitude of the detected high-frequency signal remains below the threshold if the amplitude of the detected high-frequency signal returns to a level above the threshold after reducing the amount of compression. The signal processor may be configured to detect the high-frequency signal using a phase-locked loop that monitors a signal in the feedforward signal path.

The ear cup may provide a volume surrounding the feedforward microphone, such that the screen covers an aperture between the volume surrounding the feedforward microphone and the external environment. The aperture between the volume surrounding the feedforward microphone and the external environment is at least 10 mm ² . The aperture between the volume surrounding the feedforward microphone and the external environment is at least 20 mm ² . The screen and the feedforward microphone may be separated by a distance of at least 1.5 mm.

In general, in one aspect, an active noise reducing headphone includes an earcup configured to couple to a wearer's ear to define an acoustic volume comprising a volume of air within the wearer's ear canal and a volume within the earcup, a feedback microphone acoustically coupled to the acoustic volume and electrically coupled to a feedback active noise cancellation signal path, an output transducer acoustically coupled to the acoustic volume via a first volume and electrically coupled to the feedback signal path, and a signal processor configured to apply filters and control a gain of the feedback signal path. The signal processor is configured to apply a first feedback filter to the feedback signal path and a second feedback filter to the feedback signal path, the first feedback filter causing the feedback signal path to operate at a first gain level (as a function of frequency) during a first operating mode and a second feedback filter causing the feedback signal path to operate at a second gain level less than the first gain level at certain frequencies during a second operating mode, the first gain level being a level of gain that results in effective cancellation of sound conducted through or around the earcup and through the user's head into the acoustic volume when the earcup is coupled to the wearer's ear, and the second level being a level of gain that matches the sound level of a typical wearer's voice conducted through the wearer's head when the earcup is coupled to the wearer's ear.

Various embodiments may include one or more of the following. A feedforward microphone may be acoustically coupled to an external environment and electrically coupled to a feedforward active noise cancellation signal path, such that the output transducer is electrically coupled to the feedforward signal path and the signal processor is configured to apply filters and control the gain of the feedforward signal path. In a first operating mode, the signal processor may be configured to apply a first feedforward filter to the feedforward signal path, along with a first feedback filter to the feedback signal path, to achieve effective cancellation of ambient sounds. In a second operating mode, the signal processor may be configured to apply a second feedforward filter to the feedforward signal path, the second filter selected to provide active hear-through of ambient sounds with ambient naturalness. The second feedback filter and the second feedforward filter may be selected to provide active hear-through of the user's own voice with inherent naturalness. The second feedforward filter applied to the feedforward path may have a non-minimum phase response. Sounds representative of the wearer's voice below a first frequency passively conducted through the wearer's head may be amplified when the earcup is coupled to the wearer's ear, and sounds above the first frequency may be attenuated when the earcup is so coupled, such that the feedback signal path operates over a frequency range extending above the first frequency.

The signal processor may be a first signal processor and the feedback signal path may be the first feedback signal path, wherein the headphone includes a second ear cup configured to couple to a wearer's second ear to define a second acoustic volume comprising a volume of air within the wearer's second ear canal and a volume within the second ear cup, a second feedback microphone acoustically coupled to the second acoustic volume and electrically coupled to a second feedback active noise cancellation signal path, a second output transducer acoustically coupled to the second acoustic volume via the volume within the second ear cup and electrically coupled to the second feedback active noise cancellation signal path, and a second signal processor configured to apply a filter and control the gain of the second feedback active noise cancellation signal path. The second signal processor may be configured to apply a third feedback filter to the second feedback signal path, the second feedback filter causing the second feedback signal path to operate at a first gain level during a first operating mode of the first signal processor, and to apply a fourth feedback filter to the second feedback signal path to operate at a second gain level during a second operating mode of the first signal processor. The first and second signal processors may be part of a single signal processing device. The third feedback filter may not be identical to the first feedback filter.

In general, in one aspect, a method is described for configuring active noise reducing headphones that include an earcup configured to couple to a wearer's ear to define an acoustic volume comprising a volume of air within the wearer's ear canal and a volume within the earcup, a feedforward microphone acoustically coupled to an external environment and electrically coupled to a feedforward active noise cancellation signal path, a feedback microphone acoustically coupled to the acoustic volume and electrically coupled to a feedback active noise cancellation signal path, an output transducer acoustically coupled to the acoustic volume via the volume within the earcup and electrically coupled to both the feedforward and feedback active noise cancellation signal paths, and a signal processor configured to apply filters and control gain of both the feedforward and feedback active noise cancellation signal paths. The method includes measuring, for at least one frequency, a ratio wherein an active noise reduction circuit of the earphone is inactive, wherein _Gcev is a response to ambient noise at the user's ear when the earphone is worn, and _Goev is a response to ambient noise at the user's ear when the earphone is not present, selecting a filter _Kon for the feedback path having an inverse sensitivity magnitude that causes the feedback loop to have an inverse sensitivity magnitude equal to the determined ratio at at least one frequency; selecting a filter _Kht for the feedforward signal path that will provide ambient naturalness; applying the selected filters _Kon and _Kht to the feedback and feedforward paths, respectively; measuring the ratio at at least one frequency wherein the active noise reduction circuit of the earphone is active; and modifying the phase of _Kht without changing its magnitude to minimize a deviation of the measured value from unity.

Various embodiments may include one or more of the following: selecting K _on and K _ht , applying the selected filter, and measuring the ratio may be repeated, and the phase of K _ht further adjusted, until a target balance of ambient response and own voice response is achieved. Selecting the filter for the feedforward signal path may include selecting a value of K _ht that results in a formula that is approximately equal to a predetermined target value.

In general, in one aspect, an active noise reducing headphone includes an earcup configured to couple to a wearer's ear to define an acoustic volume comprising a volume of air within the wearer's ear canal and a volume within the earcup, a feedforward microphone acoustically coupled to an external environment and electrically coupled to a feedforward active noise cancellation signal path, a feedback microphone acoustically coupled to the acoustic volume and electrically coupled to a feedback active noise cancellation signal path, a signal input for receiving an input electronic audio signal and electrically coupled to an audio playback signal path, an output transducer acoustically coupled to the acoustic volume via the volume within the earcup and electrically coupled to the feedforward and feedback active noise cancellation signal paths and the audio playback signal path, and a signal processor configured to apply filters and control gain of both the feedforward and feedback active noise cancellation signal paths. The signal processor is configured to apply a first feedforward filter to the feedforward signal path and a first feedback filter to the feedback signal path in a first operating mode that provides effective cancellation of ambient sounds, apply a second feedforward filter to the feedforward signal path during a second operating mode that provides active hear-through of ambient sounds with ambient naturalness, and provide an input electronic audio signal to the output transducer via the audio playback signal path during both the first and second operating modes.

Various embodiments may include one or more of the following. The residual sound at the ear due to external noise present in the earphone during the first operating mode may be 12 dBA less than the residual noise at the ear due to the same external noise present in the earphone during the second operating mode. The total audio level of the reproduced input audio signal of the earphone may be the same in both the first and second operating modes. The frequency response of the earphone may be the same in both the first and second operating modes, and the signal processor may be configured to change the gain applied to the audio playback signal path between the first and second operating modes. The signal processor may be configured to reduce the gain applied to the audio playback signal path during the second operating mode relative to the gain applied to the audio playback signal path during the first operating mode. The signal processor may be configured to increase the gain applied to the audio playback signal path during the second operating mode relative to the gain applied to the audio playback signal path during the first operating mode.

The headphones may include a user input that configures the signal processor to apply a second feedforward filter to the feedforward signal path during a third operating mode that provides active hear-through of ambient sound with ambient naturalness, not provide the input electronic audio signal to the output transducer via the audio playback signal path during the third operating mode, and, based on receiving a signal from the user input during the first operating mode, transition to a selected one of the second operating mode or the third operating mode. The selection of whether to transition to the second operating mode or the third operating mode may be based on a duration for which the signal from the user input is received. The selection of whether to transition to the second operating mode or the third operating mode may be based on a predetermined configuration setting of the headphones. The predetermined configuration setting of the headphones may be determined by a position of a switch. The predetermined configuration setting of the headphones may be determined by an instruction received by the headphones from a computing device. Upon entering the third processing mode, the signal processor may be configured to stop providing the input electronic audio signal by transmitting a command to the source of the input electronic audio signal to pause playback of the multimedia source.

The audio playback signal path and the output transducer may be operational when no power is applied to the signal processor. The signal processor may be further configured to disconnect the audio playback signal path from the output transducer upon activation of the signal processor and, after a time delay, reconnect the audio playback signal path to the output transducer via a filter applied by the signal processor. The signal processor may be further configured to initially maintain the audio playback signal path to the output transducer upon activation of the signal processor and, after a time delay, disconnect the audio playback signal path from the output transducer and simultaneously reconnect the audio playback signal path to the output transducer via a filter applied by the signal processor. The overall audio response of the headphone in reproducing the input audio signal when the signal processor is inactive may be characterized by a first response, and the signal processor may be configured to apply, after the time delay, an equalization filter that causes the overall audio response of the headphone in reproducing the input audio signal to remain the same as the first response, and, after a second time delay, apply a second equalization filter that results in an overall audio response that is different from the first response.

Generally, in one aspect, an active noise reducing headphone has an active noise cancellation mode and an active hear-through mode, and the headphone changes between the active noise cancellation mode and the active hear-through mode based on detecting a user touching a housing of the headphone. Generally, in another aspect, an active noise reducing headphone has an active noise cancellation mode and an active hear-through mode, and the headphone changes between the active noise cancellation mode and the active hear-through mode based on receiving a command signal from an external device.

Various embodiments may include one or more of the following. An optical detector may be used to receive the command signal. A radio frequency receiver may be used to receive the command signal. The command signal may include an audio signal. The headset may be configured to receive the command signal via a microphone integrated into the headset. The headset may be configured to receive the command signal via a signal input of the headset for receiving an input electronic audio signal.

In general, in one aspect, an active noise reducing headphone includes an earcup configured to couple to a wearer's ear to define an acoustic volume comprising a volume of air within the wearer's ear canal and a volume within the earcup, a feedforward microphone acoustically coupled to an external environment and electrically coupled to a feedforward active noise cancellation signal path, a feedback microphone acoustically coupled to the acoustic volume and electrically coupled to a feedback active noise cancellation signal path, an output transducer acoustically coupled to the acoustic volume via the volume within the earcup and electrically coupled to both the feedforward and feedback active noise cancellation signal paths, and a signal processor configured to apply filters and control gains of both the feedforward and feedback active noise cancellation signal paths. The signal processor is configured to operate the headphone in a first operating mode that provides effective cancellation of ambient sound and in a second operating mode that provides active hear-through of the ambient sound, and to change between the first and second operating modes based on a comparison of signals from the feedforward microphone and the feedback microphone.

Various embodiments may include one or more of the following. The signal processor may be configured to change from the first operating mode to the second operating mode when a comparison of the signals from the feedforward microphone and the feedback microphone indicates that the user of the headset is speaking. The signal processor may be configured to change from the second operating mode to the first operating mode after a predetermined amount of time has passed since the signals from the feedforward microphone and the feedback microphone no longer indicate that the user of the headset is speaking. The signal processor may be configured to change from the first operating mode to the second operating mode when the signal from the feedback microphone correlates with the signal from the feedforward microphone within a frequency band consistent with a portion of human speech amplified by an occlusion effect and is above a threshold level indicating that the user is speaking.

In general, in one aspect, an active noise reducing headphone has an active noise cancellation mode and an active hear-through mode, and includes an indicator that is activated when the headphone is in the active hear-through mode, the indicator being visible only from the front of the headphone at a limited viewing angle. In general, in another aspect, the active noise reducing headphone includes an earcup configured to couple to a wearer's ear to define an acoustic volume comprising a volume of air within the wearer's ear canal and a volume within the earcup, a feedforward microphone acoustically coupled to an external environment and electrically coupled to a feedforward active noise cancellation signal path, a feedback microphone acoustically coupled to the acoustic volume and electrically coupled to a feedback active noise cancellation signal path, an output transducer acoustically coupled to the acoustic volume via the volume within the earcup and electrically coupled to both the feedforward and feedback active noise cancellation signal paths, and a signal processor configured to apply filters and control the gain of both the feedforward and feedback active noise cancellation signal paths. The signal processor is configured to operate the headphone in a first operating mode that provides effective cancellation of ambient sound and in a second operating mode that provides active hear-through of ambient sound. During the second operating mode, the signal processor is configured to detect a high frequency signal in the feedforward active noise cancellation signal path exceeding a threshold level indicative of abnormally high acoustic coupling of the output transducer to the feedforward microphone, apply a compression limiter to the feedforward signal path in response to the detection, and remove the compression limiter from the feedforward signal path once the high frequency signal is no longer detected at a level above the threshold.

In general, in one aspect, active noise-reducing headphones have a noise cancellation mode and an active hear-through mode, and include a right feed-forward microphone, a left feed-forward microphone, and signal outputs for providing signals from the right and left feed-forward microphones to an external device. In general, in another aspect, a system for providing binaural telepresence includes a first communication device and a second communication device capable of receiving signals from the first communication device. A first set of active noise-reducing headphones having an active noise cancellation mode and an active hear-through mode is coupled to the first communication device and configured to provide first left and right feed-forward microphone signals to the first communication device, and a second set of active noise-reducing headphones having an active noise cancellation mode is coupled to the second communication device. The first communication device is configured to transmit the first left and right feed-forward microphone signals to the second communication device. The second communication device is configured to provide the first left and right feed-forward microphone signals to a second set of headphones. The second set of headphones is configured to activate their noise cancellation mode when reproducing the first left and right feed-forward microphone signals so that a user of the second set of headphones hears ambient noise from the environment of the first set of headphones, and to filter the first left and right feed-forward microphone signals so that the user of the second set of headphones hears the ambient noise from the first set of headphones with ambient naturalness.

Various embodiments may include one or more of the following. The second set of headphones may be configured in a first operating mode to provide a first right feed-forward microphone signal to a left ear cup of the second set of headphones, and to provide a first left feed-forward microphone signal to a right ear cup of the second set of headphones. The second set of headphones may be configured in a second operating mode to provide a first right feed-forward microphone signal to a right ear cup of the second set of headphones, and to provide a first left feed-forward microphone signal to a left ear cup of the second set of headphones. The first and second communication devices may also be configured to provide visual communication between their users, and the second set of headphones may be configured to operate in the first operating mode when visual communication is active, and to operate in the second operating mode when visual communication is not active. The first communication device may be configured to record the first left and first right feed-forward microphone signals. The second set of headphones may have an active hear-through mode and be configured to provide second left and second right feed-forward microphone signals to a second communication device, wherein the second communication device is configured to transmit the second left and second right feed-forward microphone signals to the first communication device, the first communication device being configured to provide the second left and second right feed-forward microphone signals to the first set of headphones, and the first set of headphones being configured to activate their noise cancellation mode when reproducing the second left and second right feed-forward microphone signals so that a user of the first set of headphones hears ambient noise in the environment of the second set of headphones and to filter the second left and second right feed-forward microphone signals so that the user of the first set of headphones hears the ambient noise from the second set of headphones with ambient naturalness. The first and second communication devices may be configured to coordinate the operating modes of the first and second sets of headphones so that users of both sets of headphones hear the ambient noise in the environment of a selected set of the first and second sets of headphones by placing the selected set of headphones in its active hear-through mode and placing the other set of headphones in its noise cancellation mode while reproducing the feed-forward microphone signals from the selected set of headphones.

Advantages include providing ambient and naturalness in headphones, allowing users to enjoy audio content during active hear-through mode, reducing the occlusion effect of headphones, and providing binaural telepresence.

Other features and advantages will be apparent from the description, and from the claims.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG1 shows a schematic diagram of an active noise reduction (ANR) headset.

2A to 2C illustrate the signal path through an ANR headset.

3 , 6 and 8 show block diagrams of ANR headphones with active hear-through capability.

FIG4 shows a schematic diagram of the acoustic signal path from the human throat to the inner ear.

FIG5A shows a graph of the magnitude of the occlusion effect.

FIG5B shows a graph of the insertion loss of the noise reduction circuit.

FIG7 shows a schematic diagram of a microphone housing.

DETAILED DESCRIPTION

FIG1 illustrates a typical active noise reduction (ANR) headphone system 10. A single earpiece 100 is shown; most systems include a pair of earpieces. An earcup 102 includes an output transducer or speaker 104, a feedback microphone 106 (also referred to as a system microphone), and a feedforward microphone 108. The speaker 102 divides the earcup into a front volume 110 and a rear volume 112. The system microphone 106 is typically located in the front volume 110 and is coupled to the user's ear via a cushion 114. Aspects of the configuration of the front volume of an ANR headphone are described in U.S. Patent 6,597,792, incorporated herein by reference. In some examples, the rear volume 112 is coupled to the external environment via one or more ports 116, as described in U.S. Patent 6,831,984, incorporated herein by reference. The feedforward microphone 108 is housed outside the earcup 102 and may be enclosed as described in U.S. Patent Application 2011/0044465, incorporated herein by reference. In some examples, multiple feed-forward microphones are used, and their signals are combined or used individually. References herein to a feed-forward microphone include designs having multiple feed-forward microphones.

The microphone and speaker are all coupled to the ANR circuit 118. The ANR circuit can receive additional input from a communication microphone 120 or an audio source 122. In the case of a digital ANR circuit, such as that described in U.S. Patent 8,073,150, which is incorporated herein by reference, software or configuration parameters for the ANR circuit can be obtained from a storage device 124. The ANR system is powered by a power source 126, which can be, for example, a battery, a portion of the audio source 122, or a communication system. In some examples, one or more of the ANR circuit 118, storage device 124, power source 126, external microphone 120, and audio source 122 are located within or attached to the ear cup 102, or, when two earpieces 100 are provided, are distributed between the two ear cups. In some examples, some components, such as the ANR circuit, are duplicated between the earpieces, while other components, such as the power source, are located in only one earpiece, as described in U.S. Patent 7,412,070, which is incorporated herein by reference. External noise to be cancelled by the ANR headphone system is represented as acoustic noise source 128 .

When both feedback and feedforward ANR circuits are provided in the same headphone, they are typically tuned to operate over different but complementary frequency ranges. When describing the frequency range in which the feedback or feedforward noise cancellation path operates, we are referring to the range in which ambient noise is reduced; outside this range, the noise is unchanged or may be slightly amplified. Where their operating ranges overlap, the circuit's attenuation can be intentionally reduced to avoid creating a range where cancellation is greater than elsewhere. That is, the attenuation of an ANR headphone can be modified across different frequency ranges to provide a more consistent response than would be achieved by simply maximizing stability at all frequencies or attenuating within fundamental acoustic limits. Ideally, a consistent amount of noise reduction is provided across the entire audible range between the feedback path, the feedforward path, and then the passive attenuation of the headphone. We refer to such a system as providing effective cancellation of ambient sound. To provide the active hear-through feature described below, it is ideal that the feedback path have a high-frequency crossover frequency (where attenuation drops below 0 dB) above at least 500 Hz. The feedforward loop will typically operate over a frequency range that extends above the feedback path.

This application relates to improvements in hear-through achieved through complex manipulation of active noise reduction systems. Different hear-through topologies are illustrated in Figures 2A to 2C. In the simplified version shown in Figure 2A, the ANR circuit is disabled, allowing ambient sound 200 to pass through or around the ear cup, providing passive monitoring. In the version shown in Figure 2B, the direct talk-through feature, as discussed above, uses an external microphone 120 coupled to an internal speaker 104 by the ANR circuit or some other circuitry to reproduce ambient sound directly inside the ear cup. The feedback portion of the ANR system is either left unmodified, treating the talk-through microphone signal as a normal audio signal to be reproduced, or the feedback portion of the ANR system is disabled. The talk-through signal is typically band-limited to the vocal range. For this reason, direct talk-through systems tend to sound artificial, as if the user is listening to their surroundings through a phone. In some examples, the feed-forward microphone serves dual purpose as the talk-through microphone, allowing the sound it detects to be reproduced rather than cancelled.

We define active hear-through to describe the feature of changing the active noise cancellation parameters of headphones so that the user can hear some or all of the ambient sounds in the environment. The goal of active hear-through is to allow the user to hear the environment as if they were not wearing headphones at all. That is, while direct talk-through, as in Figure 2B, tends to sound artificial, and passive listening, as in Figure 2A, obscures the ambient sounds through the passive attenuation of the headphones, active hear-through struggles to make the ambient sounds sound completely natural.

As shown in FIG2C , active hear-through (HT) is provided by using one or more feed-forward microphones 108 (only one shown) to detect ambient sound, and adjusting the ANR filter for at least the feed-forward noise cancellation loop to allow a controlled amount of ambient sound 200 to pass through the ear cup 102 with less cancellation than would otherwise be applied in normal noise cancellation (NC) operation. The ambient sound in question may include all ambient sound, just the voices of others, or the wearer's own voice.

Natural hear-through of ambient sounds.

Providing natural-sounding hear-through of ambient sounds (which we call "ambient naturalness") is accomplished by modifying the active noise cancellation filter. In systems with both feedback and feedforward noise cancellation circuits, one or both of the cancellation circuits can be modified. As explained in U.S. Patent 8,155,334 and incorporated herein, a feedforward filter implemented in a digital signal processor can be modified to provide hear-through by not completely canceling all ambient noise or a subset of the ambient noise. In an example of that application, the feedforward filter is modified to attenuate sounds less within the human voice band than outside the human voice band. The application also provides parallel analog filters as an alternative to digital filters, one for full attenuation and the other for reduced attenuation in the voice band.

To make the sound allowed to pass sound more natural, compensate for the changes in sound caused by passive attenuation, and provide natural hear-through over the entire range of audio frequencies, the feedforward filter can be modified in more complex ways. Figure 3 shows a block diagram of the ANR circuit and related components used in an example like Figure 2C. We call the effects of various components on sound moving between various points in the system responses or transfer functions. Some responses of interest are defined as follows:

a) _Goea : Response from noise to ear, without headphones

b) _Gpfb : Response from noise passing through the headphone to the ear, and feedback ANR is active

c) G _nx : Response from noise to external (feedforward) microphone

d) G _ffe : The output of the feedback filter and the response of any signal summed with it to the ear through the driver 104, and feedback ANR is active

The various electronic signal paths of the ANR circuit have the following filters applied to them, which we can refer to as the gain of the path:

_Kfb : Gain of feedback compensation filter

K _ff : Gain of the feedforward compensation filter

K _ht : Gain of the active hear-through filter (in FIG3 , K _ff and K _ht may be applied to the same path alternatively)

We define the target hear-through insertion gain as _Thtig , which represents how the overall system should filter ambient sounds. If _Thtig = 1 (0 dB), then the user should hear the world around them identically to when they are not wearing headphones. In practice, target values other than 0 dB are often desirable. For example, cancellation at low frequencies, such as below 100 Hz, is still useful during active hear-through mode, as such sounds tend to be unpleasant and lack useful information. However, a _Thtig passband that extends to cover at least the range of 300 Hz to 3 kHz is necessary to make those voices surrounding the user clearly intelligible. Preferably, the passband extends from 140 Hz to 5 kHz to achieve a sense of naturalness. The passband can be shaped to improve the perception of naturalness in active hear-through mode; for example, a slight high-frequency roll-off can compensate for the distortion of spatial hearing caused by the presence of headphones. Ultimately, the filter should be designed to provide a smooth and piecewise-linear overall system response. By "smooth and piecewise-linear," we refer to the general shape of a plot of the system response on a dB/logarithmic frequency scale.

Combining these factors, the total response to ambient noise at the ear when wearing headphones is _Gpfb + _Gnx * _Kht * _Gffe . The desired response is _Goea * _Thtig . That is, the combination of the passive and feedback responses _Gpfb and the actual hear-through response _Gnx * _Kht * _Gffe should sound like the target hear-through insertion gain _Thtig applied to the open-ear response _Goea . The system is tuned to deliver the desired response by measuring the various actual responses (those _Gxx terms) and defining the filter _Kht (within achievable limits) so that the actual system response is as close to the target as possible, based on the equation:

Solving equation (1) for K _ht yields:

To optimally achieve the desired _Thtig , the filter _Kht implemented in the feedforward signal path can be non-minimum phase, i.e., it can have zeros in the right-half plane. For example, this can allow active hear-through to pass human speech while canceling the ambient rumble present in many buildings due to heating and cooling systems. This combination is provided by designing _Kht so that _Thtig approaches 0 dB only in the active hear-through passband. Outside the active hear-through passband, _Kht is designed so that _Thtig approaches, and ideally equals, the insertion gain (actually, insertion loss) achieved by the feedforward filter (i.e., the typical _Kff ) that results in significant attenuation. The signs of the feedforward filter required for effective attenuation ( _Kff ) and active hear-through ( _Kht ) are typically opposite in sign to the hear-through passband. Designing _Kht to roll off at the low-frequency edge of the passband and transition to an effective _Kff response can be achieved by including at least one right-half-plane zero near this transition.

In general, replacing the feedforward filter _Kff with an active hear-through filter _Kht while maintaining the feedback loop _Kfb enables the ANR system to combine with the passive acoustic path through the headphones to create a natural experience at the ear that sounds the same as without headphones. To allow _Kht to transmit the intended sounds of the outside world, the feedback loop combined with the passive acoustic path through the headphones should provide at least 8dB of attenuation at all frequencies of interest. That is, the noise level heard at the ear when the feedback loop is active but the feedforward path is inactive should be at least 8dB less than the noise level at the ear when the headphones are not worn at all (note that "at least 8dB less" refers to the ratio of the levels, not a number of decibels on the same external scale). When _Gpfb is less than or equal to -8dB, its contribution to the actual hear-through insertion gain is less than a 3dB error when the desired _Thtig = 0dB. If the feedback loop can handle more gain, the attenuation can be much higher, or the passive attenuation is greater. To achieve this naturalness in some cases, as discussed below, it may also be desirable to reduce the feedback loop's gain _Kfb from its maximum capability.

The difference in total noise reduction at the ear between normal ANR mode and active hear-through mode should be at least 12 dBA. This provides a sufficient change in ambient noise level to switch from active hear-through mode with quiet background music to a dramatic change in noise reduction. This is because the perceived loudness of ambient noise decreases rapidly when switching modes due to the presence of music masking. Quiet background music in hear-through mode can render the noise virtually inaudible in noise reduction mode, as long as there is at least a 12 dBA change in noise reduction between hear-through and noise reduction modes.

In some examples, a digital signal processor, such as that described in U.S. Patent 8,184,822 and incorporated herein by reference, advantageously adds a feedback loop to the path through the feedforward microphone, avoiding the combination (deep zero in the combined signal) that can cause K _ht to have a typical delay of hundreds of microseconds typical of sound-quality ADC/DAC combinations. Preferably, the system is implemented using a DSP with a delay of less than 250 μs, so that the first potential zero of the combination (which will be 2 kHz with a 250 μs delay) is one octave higher than the typical minimum insertion loss frequency of G _pfb , which is typically around 1 kHz. The configurable processor described in the referenced patent also allows the active pass-through filter K _ht to be easily replaced with the feedforward filter K _ff .

Once ambient naturalness is achieved, additional features can be provided by selecting between more than one feedforward filter, K _ht , providing different overall response characteristics. For example, one filter may be desirable for providing hear-through in an airplane, where loud, low-frequency sounds tend to obscure conversation, so some cancellation in this frequency range should be maintained, while the voiceband signal should be passed as naturally as possible. Another filter may be desirable in a generally quieter environment where the user wants or needs to accurately hear ambient sounds, such as to provide situational awareness while walking down the street. Selecting between active hear-through modes can be accomplished using a user interface, such as a button, switch, or an app on a smartphone paired with the headphones. In some examples, the user interface for selecting the hear-through mode is a volume control, so that different hear-through filters are selected based on the volume setting selected by the user.

Hear-through filter selection can also be automatic in response to the ambient noise spectrum or level. For example, if the ambient noise is generally quiet or generally broad-spectrum, a broad-spectrum hear-through filter can be selected. However, if the ambient noise has high signal content in a particular frequency range, such as the whirring of aircraft engines or a subway train, that range may be eliminated more than would be required to provide ambient naturalness. A filter can also be selected to provide broad-spectrum hear-through, but at a reduced volume level. For example, setting T _htig = 0.5 would provide 6 dB of insertion loss across a broad frequency range. The measurement of ambient sound used to automatically select the hear-through filter can be a time-averaged measurement of the spectrum or level, which can be updated periodically or continuously. Alternatively, the measurement can be made immediately at the time the user activates hear-through mode, or a time average of the sampling times immediately before or after the user makes the selection can be used.

One example of automatic selection of an active hear-through filter is in industrial hearing protection. A headset with active noise reduction using feedback and feedforward, plus passive attenuation providing 20 dB of attenuation, can be used to protect hearing (to acceptable standards) at noise levels up to 105 dBA (i.e., it reduces 20 dB from 105 dBA to 85 dBA), which covers the majority of industrial noise pollution. However, in industrial environments where noise levels vary over time or location, a full 20 dB of attenuation is undesirable when it is relatively quiet (e.g., less than 70 dBA) because it hinders communication between workers. A multi-mode active hear-through headset can operate as a dynamic noise-reducing hearing protector. Such a device would monitor the ambient noise level at the feedforward microphone and, if the level is less than 70 dBA, apply the filter K _ht to create a feedforward path with T _htig = 0 dB. As the noise level rises above 70 dBA, the headset detects this and steps through multiple sets of K _ht filter parameters (such as from a lookup table) to gradually reduce the insertion gain. Preferably, the headset will have many possible filter sets to apply, and the detection of the ambient level will be done with a long time constant. The audible effect will be to compress the slow increase in the actual noise level around the user from 70 to 105 dBA into a perceived increase from 70 to only 85 dBA, while continuing to convey the short-term dynamics of speech and noise.

The above figures and descriptions consider a single earcup. Typically, active noise reduction headphones have two earcups. In some examples, the same hear-through filter is applied to both earcups, but in other examples, different filters may be applied, or the hear-through filter K _ht may be applied to only one earcup while the feedforward cancellation filter K _ff is maintained in the other earcup. This can be advantageous in various examples. If the headphones are used by a driver communicating with other vehicles or a control center, enabling hear-through in only one earcup can allow the driver to speak with crew members who are not wearing the headphones, while maintaining awareness of communication signals or warnings by maintaining noise cancellation activity in the other earcup.

Active hear-through performance can be enhanced if the feed-forward microphone signal for each ear cup is shared with the other ear cup and inserted into the signal path of each opposing ear cup using another set of filters K _xo . This can provide directionality to the hear-through signal, so the wearer can better determine the source of sounds in their environment. Such an improvement can also increase the perceived relative level of a person's voice coaxially in front of the wearer relative to diffuse ambient noise. A system capable of providing cross-feed-forward signals is described in U.S. Patent Application Publication No. 2010/0272280, incorporated herein by reference.

In addition to using active noise cancellation technology to provide both ANR and hear-through, an active hear-through system can also include a single-channel noise reduction filter in the feed-forward signal path during hear-through mode. Such a filter can clean up the hear-through signal, for example, improving speech intelligibility. Such in-channel noise reduction filters are known for use in communication headsets. For optimal performance, such filters should be implemented within the latency constraints described above.

When a feedforward microphone is used to provide active hear-through of ambient sound, protecting the microphone from wind noise (i.e., noise caused by air moving rapidly past the microphone) can be beneficial. Headphones used indoors, such as on aircraft, generally do not require wind noise protection, but headphones that may be used outdoors may be susceptible. As schematically shown in FIG7 , an effective way to protect the feedforward microphone 108 from wind noise is to provide a screen 302 over the microphone and some distance between the screen and the microphone. Specifically, the distance between the screen and the microphone should be at least 1.5 mm, while the hole in the earcup housing 304 covered by the screen 302 should be as large as possible. Given the practical considerations of fitting such a component in an in-ear headphone, the screen area should be at least 10 ^mm² , preferably 20 ^mm² or greater. The total volume enclosed by the screen and the sidewalls 306 of the cavity 308 surrounding the microphone 108 is not critical, so the space around the microphone can be tapered with the microphone at the apex and the angle of the taper selected to provide as large a screen area as other packaging constraints allow. The screen should have some considerable acoustic resistance, but not so much as to reduce the sensitivity of the microphone to an unnecessarily low level. Acoustically resistive cloth with a specific acoustic resistance between 20 and 260 Rayls (mkS) has been found to be effective. If the headset is to be used in a windy environment, such protection can also be valuable for general noise reduction by preventing wind noise from saturating the feedforward cancellation path.

Natural hearing of the user's voice

When a person hears their own voice sound natural, we call it "self-naturalness." As just described, ambient naturalness is achieved by modifying the feedforward filter. Self-naturalness is provided by modifying the feedforward filter and the feedback system, but the changes are not necessarily the same as those used when ambient naturalness itself is desired. Typically, achieving both ambient naturalness and self-naturalness in active hear-through requires changes to both the feedforward and feedback filters.

As shown in Figure 4, a person typically hears their own voice through three acoustic pathways. The first pathway 402 is through the air near the head 400, from the mouth 404 to the ear 406 and into the ear canal 408 to reach the eardrum 410. In the second pathway 412, acoustic energy travels through the soft tissue 414 of the neck and head, from the throat 416 to the ear canal 408. The sound then enters the air volume inside the ear canal through vibrations of the ear canal walls, joining the first pathway to reach the eardrum 410, but also exiting through the ear canal opening into the air outside the head. Finally, in the third pathway 420, the sound also travels from the throat 416 through the soft tissue 414, as well as through the Eustachian tube connecting the throat to the middle ear 422. It then enters the middle ear 422 and inner ear 424 directly, bypassing the ear canal and joining the sound entering through the eardrum from the first two pathways. In addition to providing different signal levels, the three pathways contribute different frequency components to what the user hears as their own voice. The second path 412 through the soft tissue to the ear canal is the dominant body-conducted path at frequencies below 1.5 kHz and can be as important as the air-conducted path at the lowest frequencies of the human voice. Above 1.5 kHz, the third path 420 directly to the middle and inner ear is dominant.

When wearing headphones, first path 402 is blocked to some extent, so the user cannot hear portions of their own voice, changing the mix of signals reaching the inner ear. In addition to the contribution from the second path providing a larger share of the total sound energy reaching the inner ear due to losses in the first path, the second path itself becomes more efficient when the ear is blocked. When the ear is open, sound that enters the ear canal via the second path can exit the ear canal through the ear canal opening. Blocking the ear canal opening improves the efficiency of coupling ear canal wall vibrations into the air in the ear canal, which increases the amplitude of pressure oscillations in the ear canal and, in turn, increases the pressure on the eardrum. This is commonly known as the occlusion effect, and it can amplify sounds at the fundamental frequency of a male voice by as much as 20-25 dB. As a result of these changes, the user perceives their voice as having overemphasized lower frequencies and underemphasized higher frequencies. In addition to making the voice sound lower, the removal of higher-frequency sounds from the human voice also makes the voice less intelligible. This change in the user's perception of their own voice can be addressed by modifying the feedforward filter to allow the air-conducted portion of the user's voice, and modifying the feedback filter to counteract the occlusion effect. As discussed above, if the occlusion effect can be reduced, changes to the feedforward filter for ambient naturalness are often sufficient to also provide self-naturalness. Reducing the occlusion effect can have benefits beyond self-naturalness, and will be discussed in more detail below.

Reduction of occlusion effect

The occlusion effect is particularly strong when the earphones are first put on, that is, by directly blocking the entrance to the ear canal but not extending far into the ear canal. Larger ear cups provide more space for sound to exit the ear canal and dissipate, and deep-canal earpieces block some of the sound from initially passing through the soft tissue into the ear canal. If the earphones or earbuds extend far enough into the ear canal, past the muscle and cartilage to the very thin skin over the bone of the skull, the occlusion effect is mitigated because little sound pressure enters the enclosed volume through the bone, but extending the earphones that far into the ear canal is difficult, dangerous, and potentially painful. For any type of earphone, creating any amount of occlusion effect reduction is beneficial for providing naturalness in active hear-through features and for removing non-vocal elements of the occlusion effect.

The experience of wearing headphones is improved by eliminating the occlusion effect, so that the user can hear their own voice naturally when active hear-through is provided. Figure 6 shows a schematic diagram of the headphone system and the various signal paths passing through it. The external noise source 200 and the associated signal paths in Figure 3 are not shown, but may appear in conjunction with the user's voice. The feedback system microphone 106 and the compensation filter _Kfb create a feedback loop that detects and eliminates the sound pressure within the volume 502 limited by the headphone 102, the ear canal 408, and the eardrum 410. This is the same volume where the amplified sound pressure that causes the occlusion effect at the end of the path 412 exists. As a result of the feedback loop that reduces the amplitude of the oscillations in this pressure (i.e., sound), the oscillation effect is reduced or eliminated by the normal operation of the feedback system.

Reducing or even eliminating the negative effects of the ringing effect can be accomplished without completely eliminating the sound pressure. Some feedback-canceling headphones provide more cancellation than necessary to mitigate the occlusion effect. When the goal is simply to remove the occlusion effect, the feedback filter or gain is adjusted to provide just enough cancellation to remove the occlusion effect without further eliminating ambient sounds. We denote this by applying filter K _on in place of the full feedback filter K _fb .

As shown in FIG5A , the occlusion effect is most pronounced at low frequencies and decreases as frequency increases, becoming imperceptible (0 dB) somewhere in the mid-frequency range between 500 Hz and 1500 Hz, depending on the specific design of the headphones. The two examples in FIG5A are over-ear headphones (curve 452), for which the occlusion effect ends at 500 Hz, and in-ear headphones (curve 454), for which the occlusion effect extends to 1500 Hz. Feedback ANR systems are typically effective in the low to mid-frequency range (i.e., they can reduce noise), losing their effectiveness somewhere in the same range where the occlusion effect ends, as shown in FIG5B . In the example of FIG5B , the insertion loss (i.e., the reduction in sound from the outside of the ear cup to the inside) curve 456 crosses above 0 dB near 10 Hz and crosses back below 0 dB near 500 Hz due to the ANR circuit. If the feedback ANR system in a given headphone is effective up to a frequency above the frequency at which its occlusion effect terminates in that headphone, such as curve 452 in FIG3 , the magnitude of the feedback filter can be reduced and still remove the occlusion effect overall. On the other hand, if the feedback ANR system stops providing effective noise reduction at a frequency below the frequency at which its occlusion effect terminates for that headphone, such as curve 454 in FIG5A , then the full magnitude of the feedback filter will be required and some occlusion effect will remain.

As with the feedforward system, the filter parameters for the feedback system that achieve naturalness by eliminating occlusion effects as much as possible can be found from the responses of the various signal paths in the headphone system shown in Figure 6. In addition to the same ones as in Figure 3, the following responses are also considered:

a) G _ac : Response of the air conduction path 402 from mouth to ear (not blocked by earphones, as shown in FIG4 )

b) G _bcc : Response of the body conduction path 412 to the ear canal (when the ear canal is not blocked by the earphone)

c) G _bcm : Response of the body conduction path 420 to the middle ear and inner ear

The body-conducted responses _Gbcc and _Gbcm are significant at different frequency ranges, typically below and above 1.5kHz, respectively. These three paths combine to form the net open-ear response of the user's voice at the ear canal without an earphone, G _oev = G _ac + G _bcc + G _bcm . Conversely, the net closed-ear voice response when an earphone is present is defined as G _cev .

The net response, G _oev or G _{cev ,} cannot be measured directly with any repeatability or precision, but their ratio, G _cev /G _{oev ,} can be measured by suspending a miniature microphone in the ear canal (without blocking the ear canal) and finding the ratio of the spectrum measured when the subject speaks while wearing the earphones to the spectrum measured when the subject speaks without wearing the earphones. Performing the measurement on both ears, one blocked by the earphones and the other open, prevents errors caused by variability in human speech between measurements. Such a measurement is the source of the occlusion effect curve in Figure 5A.

To find a value of K _on to use to cancel out just the occlusion effect, we consider the contribution of the headphone and the ANR system as they combine to form _Gcev on the response. A reasonable estimate is that _Gac is affected in the same way as air-conducted ambient noise, so its contribution to _Gcev is _Gac *( _Gpfb + _Gnx * _Kht * _Gffe ). The headphone has a negligible effect on the third path directly to the middle and inner ear, so _Gbcm remains unchanged. For the second path 412, the body-conducted sound entering the ear canal is indistinguishable from the ambient noise passing through the ear cup, so the feedback ANR system cancels it using the feedback loop occlusion filter _Kon , providing a response of _Gbcc /(1- _Lfb ), where the loop gain _Lfb is the product of the feedback filter _Kon and the driver-to-system microphone response _Gds . Overall, then,

as well as

For self-naturalness, it is desirable that _Gcev / _Goev = 1 (0 dB). Combined with the earlier equation (1) for self-naturalness, this allows balancing these two aspects of the hear-through experience. Human perception of ambient sound is very insensitive to phase (assuming the phase does not change very rapidly), so the value of _Kht chosen to estimate _Thtig results in an insignificant phase response. The key to solving equation (1) for _Kht is matching the magnitudes | _Thtig |. However, the phase of _Gpfb + _Gnx * _Kht * _Gffe affects how the covered ear _Gac path (affected by _Kht ) adds to the covered ear _Gbcc path (affected by _Kon ). The design process breaks down into the following steps:

a) Measure the occlusion effect (low frequency boost in _Gcev / _Goev ) by measuring _Gcev with all ANR turned off.

b) Design the ANR feedback loop to balance the measured occlusion effect. If measurements show a 10 dB boost in occlusion at 400 Hz, then as a first guess one would want a 10 dB inverse sensitivity of the feedback loop at that frequency (1-Lfb). For headphones that do not have enough feedback ANR gain to completely eliminate the occlusion effect, _Kon will simply be equal to the optimized feedback loop's _Kfb . For headphones with sufficient headroom in the feedback loop, _Kon will be some value less than _Kfb .

c) Design K _ht for the naturalness of the environment as described above.

d) Apply a K _ht filter to the feedforward loop and K _on to the feedback loop and measure G _cev /G _oev again.

e) Adjust the phase of K _ht without changing the magnitude by adding all-pass filter stages or moving zeros into the right half plane, so as to minimize any deviation of G _cev /G _oev from 1 (transparency).

f) It may also be advantageous to adjust K _on during this process. The updated values of K _on and K _ht are repeated to find the best balance of the desired ambient response and own voice response.

Reducing the occlusion effect and allowing the wearer to naturally hear their own voice has the further benefit of encouraging the user to speak at a normal volume when addressing another person. When people are listening to music or other sounds on headphones, they tend to speak too loudly because they speak loud enough to be heard above other sounds, even if no one else can hear them. Conversely, when people are wearing noise-canceling headphones but not listening to music, they tend to speak too softly to be understood by others in noisy environments, apparently because in this case they can easily hear their own voice above the quiet residual ambient noise they hear. The way people adjust the volume of their own speech in response to how they hear their own voice relative to other ambient sounds is called the Lombard reflex. Allowing a user to accurately hear the volume of their own voice via active hear-through allows them to properly control that volume. If playing music in the headphones causes the user to speak too loudly, reducing the music when switching to hear-through mode can also help the user properly hear their own voice and control its volume.

Maintain entertainment audio during active hear-through

Headphones that provide direct talk-through or passive listening via attenuated ANR circuitry and either reproduce external sounds or allow external sounds to passively move through the headphones also attenuate any input audio they may be reproducing, such as music. In the system described above, active noise reduction and active hear-through can be independently provided to reproduce entertainment audio. FIG8 shows a block diagram similar to that in FIG3 and FIG5 , modified to also illustrate the audio input signal path. For clarity, external noise and associated acoustic signals are not shown. In the example of FIG8 , an audio input source 800 is connected to a signal processor, filtered by an equalizing signal filter K _eq , and combined with feedback and feedforward signal paths to be delivered to the output transducer 104. The connection between source 800 and the signal processor can be a wired connection via connectors on the ear cups or elsewhere, or it can be a wireless connection using any available wireless interface, such as Wi-Fi or proprietary RF or IR communication.

Providing a separate path for input audio allows the headphones to be configured to adjust active ANR to provide active hear-through while simultaneously maintaining entertainment audio playback. Input audio can be played at some reduced volume or maintained at full volume. This allows users to interact with each other, such as flight attendants, without missing out on whatever they are listening to, such as dialogue from a movie. Additionally, it allows users to listen to music without being isolated from their surroundings, if that is desired. This allows users to wear headphones for background listening while maintaining situational awareness and staying connected to their environment. Situational awareness can be valuable, for example, if someone walking down the street in an urban setting wants to be aware of the people and traffic around them but may also want to listen to music to enhance their mood or podcasts or radio for information. They may even wear headphones to send a social "do not disturb" signal while actually wanting to be aware of what is happening around them. Even if situational awareness is not valuable, for example, if a user is listening to music at home free of other distractions, some users may want to be aware of their surroundings without the isolation that even passive headphones typically provide. Maintaining active hear-through while listening to music provides this experience.

The characteristics of the feedforward and input audio signal path filters will affect how active hear-through interacts with the reproduction of the input audio signal to produce the overall system response. In some examples, the system is tuned so that the overall audio response is the same in both noise cancellation and active hear-through modes. That is, the sound reproduced by the input audio signal sounds the same in both modes. If K _on ≠ K _fb , then K _eq must be different in the two modes by changing the inverse sensitivity from 1-G _ds K _fb to 1-G _ds K _on . In some examples, the frequency response remains the same, but the gain applied to the input audio and feedforward paths is modified. In one example, the gain of K _eq is reduced during active hear-through mode, reducing the output volume of the input audio. This can have the effect of keeping the overall output volume constant between active noise cancellation mode, where the input audio is the only audible audio, and hear-through mode, where the input audio is combined with ambient noise.

In another example, the gain of K _eq is increased during active hear-through mode, causing the output volume of the input audio to be increased. Increasing the volume of the input audio signal reduces the extent to which ambient noise inserted during active hear-through masks the input audio signal. This can have the effect of preserving the intelligibility of the input audio signal by keeping it louder than the background noise, which of course is increased during active hear-through mode. Of course, if it is desired to attenuate the input audio during active hear-through mode, this can be accomplished by simply setting the gain of K _eq to zero or by shutting down the input audio signal path (which can be the same thing in some embodiments).

Providing ANR and audio playback through separate signal paths also allows audio playback to be maintained even when the ANR circuit is completely unpowered, either because the user has turned it off or because the power supply is unavailable or depleted. In some examples, a second audio path with a different equalization filter _Knp implemented in a passive circuit is used to deliver the input audio signal to the output transducer, bypassing the signal processor. The passive filter _Knp can be designed to reproduce the system response experienced when the system is powered as closely as possible without unduly compromising sensitivity. When such a circuit is available, the signal processor or other active electronics will disconnect the passive path and replace it with the active input signal path when the active system is powered. In some examples, the system can be configured to delay reconnection of the input signal path because the active system is now operating the signal to the user. The active system can also fade in the input audio signal upon powering on, both as a signal to the user that it is operating and to provide a smoother transition. Alternatively, the active system can be configured to make the transition from passive to active audio as smooth as possible, without any drop in the audio signal. This can be accomplished by maintaining the passive signal path until the active system is ready to take over, applying integration of filters to match the active signal path to the passive path, switching from the passive path to the active path, and then fading in the desired active _Keq filter.

When active hear-through and audio reproduction are enabled simultaneously, the user interface becomes more complex than in typical ANR headphones. In one example, audio is retained by default during active hear-through, and a sequential switch, pressed to toggle between noise reduction and hear-through modes, is held to additionally mute the audio when hear-through is activated. In another example, the choice of whether to mute the audio when hear-through is engaged is a setting configured in the headphones based on the user's preference. In another example, when active hear-through is enabled, headphones configured to control a playback device, such as a smartphone, can signal the device to pause audio playback instead of mute the audio within the headphones. In the same example, such headphones can be configured to activate active hear-through mode whenever music is paused.

Other UI considerations

Typically, headphones with an active hear-through feature will include some user controls for activating the feature, such as buttons or switches. In some examples, this user interface can take the form of a more complex interface, such as an accelerometer or capacitive sensor in the earcup that detects when the user touches the earcup in a specific way that is interpreted as requesting active hear-through mode. In some cases, additional controls are provided. For situations where someone other than the user might need to activate hear-through mode, such as a flight attendant requiring a passenger's attention or a teacher requiring a student's attention, an external remote control may be desirable. This can be implemented using any conventional remote control technology, but there are some considerations due to the likely use cases of such a device.

In an aircraft, it's assumed that multiple passengers are wearing compatible headphones, but their selection of these products has not been coordinated with each other or the airline, leaving flight attendants without the information, such as unique device IDs, needed to specify which headphones should have their hear-through mode activated. In this case, it might be desirable to provide a line-of-sight remote control, such as an infrared control with a narrow beam, that must be aimed directly at a given set of headphones to activate their hear-through mode. However, in another scenario, such as during a pre-flight announcement or in an emergency, the crew might need to activate hear-through on all compatible headphones. For this scenario, multiple wide-beam infrared emitters could be placed throughout the aircraft, positioned to ensure every seat is covered. Another source of remote control suitable for aircraft use cases is to superimpose a control signal on the audio input line. In this approach, any set of headphones plugged into the aircraft's entertainment audio can be provided with a signal, providing both an announcement and seat-specific signaling. In a classroom, military, or commercial environment, on the other hand, it may be the case that all headsets are purchased or at least coordinated by a single entity, so a unique device identifier may be available and a broadcast type of remote control such as a radio can be used to turn active hear-through on and off at an individual's specific headset.

Headphones with active circuitry typically include a visible indicator of their status, often a simple on/off light. When providing active hear-through, additional indicators are beneficial. At the simplest level, a second light can indicate to the user that active hear-through mode is active. Additional indicators can be valuable for situations where the user might use active hear-through mode to communicate with others, such as flight crew or colleagues in an office environment. In some examples, a light visible to others illuminates red when ANR is active but active hear-through is inactive, and changes to green when active hear-through is active, indicating to others that they can now speak to the user of the headphones. In some examples, the indicator light is structured so that it is visible only from a narrow range of angles, such as directly in front of the user, so that only someone actually facing the user will be aware of the status their headphones are in. This allows the wearer to still use the headphones, while socially sending a "do not disturb" signal to others they are not facing.

Automatic hear-through while speaking

In some examples, the feedback system is also used to automatically enable active hear-through. When a user begins speaking, as described above, the amplitude of low-frequency pressure variations inside their ear canal increases due to the sound pressure traveling from the throat to the ear canal through the soft tissues. The feedback microphone detects this increase. In addition to eliminating the increased pressure as part of compensating for the occlusion effect, the system can use this increase in pressure amplitude to identify that the user is speaking and, therefore, enable full active hear-through mode to provide the naturalness of the user's voice. Bandpass filtering the feedback microphone signal, or correlation between the feedback and feedforward microphone signals, can be used to ensure that active hear-through is enabled only in response to the voice and not other internal pressure sources such as blood flow or body movement. When the user is speaking, both the feedforward and feedback microphones detect the user's voice. The feedforward microphone detects the air-conducted portion of the user's voice, which may cover the entire frequency range of human speech, while the feedback microphone detects the portion of the voice transmitted through the head that happens to be amplified by the occlusion effect. The envelopes of these signals are thus correlated within the band amplified by the occlusion effect when the user is speaking. If another person is speaking close to the user, the feedforward microphone may detect signals similar to those of the user speaking, while any residual sound of the speech detected by the feedback microphone will be significantly lower in volume. By checking the correlation and volume of the signal for values consistent with the user speaking, the headphones can determine when the user is speaking and activate the active hear-through system accordingly.

In addition to allowing the user to naturally hear his own voice, the automatic activation of the active hear-through feature also allows the user to hear the response of anyone he is speaking to. In such an example, after the user stops speaking, the hear-through mode can be maintained for some amount of time.

Automatic active hear-through mode is also beneficial when the headset is connected to a communication device, such as a wireless phone, that does not provide sidetone (i.e., a reproduction of the user's own voice on the near-end output). By turning on hear-through when the user is speaking or when the headset electronically detects that a call is in progress, the user naturally hears their own voice and will speak into the phone at an appropriate volume. If the communication microphone is part of the same headset, the correlation between the microphone signal and the feedback microphone signal can be used to further confirm that the user is speaking.

Stability protection

The active hear-through feature has the potential to introduce new failure modes in ANR headphones. If the output transducer is acoustically coupled to the feedforward microphone to a greater extent than would be the case under normal operation, a positive feedback loop can be created, resulting in high-frequency ringing that can be unpleasant or annoying to the user. This can occur, for example, if the user cups their hands over their ears while using headphones with a back cavity that is terminated or open to the environment, or if the headphones are removed from the head while the active hear-through system is activated, allowing free-space coupling from the front of the output transducer to the feedforward microphone.

This risk can be mitigated by detecting high-frequency signals in the feedforward signal path and activating the compression limiter if those signals exceed a volume or amplitude threshold indicating the presence of such a positive feedback loop. Once the feedback is eliminated, the limiter can be deactivated. In some examples, the limiter is gradually deactivated, and if feedback is detected again, it is raised back to the lowest level at which no feedback was detected. In some examples, a phase-locked loop monitoring the output _Kff of the feedforward compensator is configured to lock onto a relatively pure tone over a predefined frequency span. When the phase-locked loop achieves lock, this indicates instability that will subsequently trigger the compressor along the feedforward signal path. The gain at the compressor is reduced at a prescribed rate until the gain is low enough to stop the oscillatory condition. When the oscillation stops, the phase-locked loop loses lock and releases the compressor, allowing the gain to return to normal operating values. Because oscillation must first occur before it can be suppressed by the compressor, if the physical condition (e.g., hand position) persists, the user will hear a repetitive chirping sound. However, a continuous, loud chirping sound is much more unpleasant than a short, repetitive, quiet chirp.

Binaural Remote Presence

Another feature made possible by the availability of active hear-through is shared binaural telepresence. For this feature, the feedforward microphone signals from the right and left earcups of a first set of headphones are transmitted to a second set of headphones, which reproduces them using its own equalization filters based on the acoustic characteristics of the second set of headphones. The transmitted signals can be filtered to compensate for the specific frequency response of the feedforward microphones, providing a more normalized signal to the remote headphones. Playing back the feedforward microphone signals of the first set of headphones in the second set of headphones allows the user of the second set of headphones to hear the environment of the first set of headphones. This arrangement can be reciprocal, so that both sets of headphones transmit their feedforward microphone signals to each other. The users can choose to each hear the other's environment, or they can both choose to hear a single environment. In the latter mode, the two users "share" the source user's ears, and the remote user can choose to engage in full noise cancellation mode to be immersed in the source user's acoustic environment.

Such a feature can make simple communication between two people more immersive and may also have industrial applications, such as allowing a remote technician to hear the environment of a local colleague or client attempting to design or diagnose an audio system or problem in a facility. For example, an audio system engineer installing the audio system in a new auditorium may wish to consult with another system engineer located in their home office about the sound produced by the audio system. By having both parties wear such headphones, the remote engineer can hear what the installer hears with sufficient clarity to provide quality, simplified instructions on how to debug the system, thanks to the active hear-through filter.

Such a binaural telepresence system requires some system for communication, and a way to provide microphone signals to the communication system. In one example, a smartphone or tablet can be used. At least one set of headphones that provide remote audio signals is modified from a conventional design to provide feed-forward microphone signals for both ears as output to the communication device. Headphone audio connections for smartphones and computers typically include only three signal paths - stereo audio to the headphones, and mono microphone audio from the headphones to the phone or computer. The binaural output from the headphones in addition to any communication microphone output can be accomplished through non-standard applications of existing protocols, such as by making the headphones operate as a Bluetooth stereo audio source and a telephone receiver (contrary to conventional arrangements). Alternatively, the additional audio signals can be provided by a wired connection with more connectors than a typical headphone jack, or a proprietary wireless or wired digital protocol can be used.

The signals are then delivered to the communication device, which then transmits the audio signal pair to the remote communication device, which provides them to the second earphone. In the simplest configuration, the two audio signals can be delivered to the receiving earphone as standard stereo audio signals, but it may be more efficient to deliver them separately to the earphone from a common stereo audio input.

If the communication device used in this system also provides video conferencing, allowing users to see each other, flipping the left and right feedforward microphone signals can also be desirable. This way, if one user responds to a sound to their left, the other user hears that sound in their right ear, matching the direction of the remote user looking at the video conferencing display. This signal preservation can be accomplished at any point in the system, but it is likely most efficient if done by the receiving communication device, as that device will be aware of whether the user at that end is receiving the video conferencing signal.

Another feature made possible by providing the feedforward microphone signal as an output from the headphones is binaural recording with the ambient naturalness of the playback. That is, a binaural recording made using the original or microphone-filtered signal from the feedforward microphone can be played back using the K _eq of the playback headphones, so that the person listening to the recording feels fully immersed in the original environment.

Other implementations are within the scope of the following claims and other claims that the applicant may issue.

Claims (32)

1. An active noise-canceling headphone, comprising: An earpiece configured to couple to a wearer’s ear to define a sound volume, the sound volume including the volume of air within the wearer’s ear canal and the volume within the earpiece; A feedforward microphone, which is acoustically coupled to the external environment and electrically coupled to a feedforward active noise cancellation signal path; A feedback microphone, which is acoustically coupled to the acoustic volume and electrically coupled to the feedback active noise cancellation signal path; An output transducer, which is acoustically coupled to the acoustic volume via the volume within the earpiece and electrically coupled to both the feedforward active noise cancellation signal path and the feedback active noise cancellation signal path; and A signal processor configured to apply a filter and control the gain of both the feedforward active noise cancellation signal path and the feedback active noise cancellation signal path; The signal processor is configured as follows: During the first operating mode that provides effective cancellation of ambient sound, a first feedforward filter is applied to the feedforward signal path and a first feedback filter is applied to the feedback signal path, and During the second operating mode, which provides active hearing of ambient sound with environmental naturalness, the first feedforward filter is replaced by a second feedforward filter in the feedforward signal path. Furthermore, the second feedforward filter is selected to have a value K _{_ht} that causes the formula to be equal to the predetermined target response. Where G _{_pfb} is the transfer function from external noise to the ear via the headphones when the feedback active noise cancellation signal path is active; G _{_oea} is the transfer function from external noise to the ear when the headphones are not active; G _{_nx} is the transfer function from external noise to the feedforward microphone; and G _{_ffe} is the transfer function of the filtered signal through the output transducer to the ear when the feedback active noise cancellation signal path is active. The predetermined target response is a target transparency insertion gain ( _Thtig ), which is 0 dB in the transparency passband and is the same as the insertion gain achieved by the first feedforward filter outside the transparency passband. 2. The headphones of claim 1, wherein the second feedforward filter causes the headphones to have a smooth and piecewise linear overall system response at the wearer's ear. 3. The headphones according to claim 1, wherein the difference in total noise reduction of speech noise between the first operating mode and the second operating mode is at least 12 dBA. 4. The earphone of claim 1, wherein the signal processor is further configured to apply a second feedback filter, different from the first feedback filter, to the feedback signal path during the second operating mode. 5. The earphone of claim 1, wherein the combination of the feedback signal path and the earpiece reduces ambient noise at the entrance to the ear canal by at least 8 dB at all frequencies between 100 Hz and 10 kHz. 6. The earphone of claim 1, wherein the feedback signal path operates over a frequency range extending above 500 Hz. 7. The headphones of claim 2, wherein the second feedforward filter causes the overall system response to be smooth and piecewise linear in the region extending above 3 kHz. 8. The headphones of claim 7, wherein the second feedforward filter causes the overall system response to be smooth and piecewise linear in the region extending to frequencies below 300 Hz. 9. The earphone according to claim 1, wherein the feedback signal path is implemented in a digital signal processor and has a delay of less than 250 μs. 10. The earphone of claim 1, wherein the second feedforward filter is defined at a non-minimum phase zero in the transfer function that characterizes the feedforward signal path. 11. The earphone according to claim 1, wherein the signal processor is further configured to: A third feedforward filter is applied to the feedforward signal path during a third operating mode that provides active hearing of ambient sound with a total response different from that provided in the second operating mode. 12. The earphone of claim 11, further comprising user input, and wherein the signal processor is further configured to select between the first feedforward filter, the second feedforward filter, or the third feedforward filter based on the user input. 13. The headphones of claim 12, wherein the user input includes volume control. 14. The earphone of claim 11, wherein the signal processor is configured to automatically select between the second feedforward filter and the third feedforward filter. 15. The earphone of claim 14, wherein the signal processor is configured to select between the second feedforward filter and the third feedforward filter based on a time-averaged measurement of the level of the ambient noise. 16. The earphone of claim 15, wherein the signal processor is configured to select between the second feedforward filter and the third feedforward filter when a user input call for activating a transparent mode is received. 17. The earphone of claim 15, wherein the signal processor is configured to periodically select between the second feedforward filter and the third feedforward filter. 18. The earphone of claim 1, wherein the signal processor is a first signal processor and the feedforward signal path is a first feedforward signal path, the earphone further comprising: A second earpiece, configured to be coupled to the wearer’s second ear to define a second acoustic volume, the second acoustic volume including the volume of air within the wearer’s second ear canal and the volume within the second earpiece; A second feedforward microphone is acoustically coupled to the external environment and electrically coupled to a second feedforward active noise cancellation signal path. The second feedback microphone is acoustically coupled to the second acoustic volume and electrically coupled to the second feedback active noise cancellation signal path. A second output transducer, which is acoustically coupled to the second acoustic volume via the volume within the second earpiece, and electrically coupled to both the second feedforward active noise cancellation signal path and the second feedback active noise cancellation signal path; and A second signal processor is configured to apply a filter and control the gain of both the second feedforward active noise cancellation signal path and the second feedback active noise cancellation signal path. The second signal processor is configured as follows: During the first operating mode of the first signal processor, a third feedforward filter is applied to the second feedforward signal path and the first feedback filter is applied to the second feedback signal path. A fourth feedforward filter is applied to the second feedforward signal path during the second operating mode of the first signal processor. 19. The earphone of claim 18, wherein the first signal processor and the second signal processor are part of a single signal processing device. 20. The earphone according to claim 18, wherein the third feedforward filter is different from the first feedforward filter. 21. The earphone of claim 18, wherein only one of the first signal processor or the second signal processor applies a corresponding second feedforward filter or the fourth feedforward filter to the corresponding first feedforward signal path or the second feedforward signal path during a third operating mode. 22. The headphones of claim 21, further comprising user input, wherein the third operating mode is activated in response to the user input. 23. The earphone of claim 18, wherein the first signal processor is further configured to: Receive cross signals from the second feedforward microphone. A fifth feedforward filter is applied to the cross signal, and The filtered cross signal is inserted into the first feedforward signal path. 24. The earphone of claim 1, wherein the signal processor is further configured to apply a single-channel noise reduction filter to the first feedforward signal path during the second operating mode. 25. The earphone according to claim 1, wherein the signal processor is further configured to: Detecting high-frequency signals in the feedforward signal path, The amplitude of the detected high-frequency signal is compared with a threshold indicating the positive feedback loop, and If the amplitude of the detected high-frequency signal is higher than the threshold, the compression limiter is activated. 26. The earphone of claim 25, wherein the signal processor is further configured to: When the amplitude of the detected high-frequency signal is no longer higher than the threshold, the compression applied by the limiter is gradually reduced, and If, after reducing the compression amount, the amplitude of the detected high-frequency signal returns to a level above the threshold, then the compression amount is increased to a minimum level, and the amplitude of the detected high-frequency signal remains below the threshold at the minimum level. 27. The earphone of claim 25, wherein the signal processor is configured to detect the high-frequency signal using a phase-locked loop that monitors the signal in the feedforward signal path. 28. The earphone of claim 1, wherein the earpiece provides a volume surrounding the feedforward microphone, and It further includes a mesh screen covering the opening between the volume surrounding the feedforward microphone and the external environment. 29. The earphone of claim 28, wherein the aperture between the volume surrounding the feedforward microphone and the external environment is at least 10 ^mm² . 30. The earphone of claim 28, wherein the aperture between the volume surrounding the feedforward microphone and the external environment is at least 20 ^mm² . 31. The headphones of claim 28, wherein the mesh screen and the feedforward microphone are separated by a distance of at least 1.5 mm. 32. A method of operating active noise-canceling headphones, the headphones comprising: An earpiece configured to couple to a wearer’s ear to define a sound volume, the sound volume including the volume of air within the wearer’s ear canal and the volume within the earpiece; A feedforward microphone, which is acoustically coupled to the external environment and electrically coupled to a feedforward active noise cancellation signal path; A feedback microphone, which is acoustically coupled to the acoustic volume and electrically coupled to the feedback active noise cancellation signal path; An output transducer, which is acoustically coupled to the acoustic volume via the volume within the earpiece and electrically coupled to both the feedforward active noise cancellation signal path and the feedback active noise cancellation signal path; and A signal processor configured to apply a filter and control the gain of both the feedforward active noise cancellation signal path and the feedback active noise cancellation signal path. The method includes: In the signal processor, during a first operating mode, effective cancellation of ambient noise is provided by applying a first feedforward filter to the feedforward signal path and a first feedback filter to the feedback signal path. In the signal processor, during the second operating mode, active hearing of ambient sound with environmental naturalness is provided by replacing the first feedforward filter with a second feedforward filter and applying the second feedforward filter to the feedforward signal path. The second feedforward filter has a value K _{_ht} selected to make the formula equal to the predetermined target response. Where G _{_pfb} is the transfer function from external noise to the ear via the headphones when the feedback active noise cancellation signal path is active; G _{_oea} is the transfer function from external noise to the ear when the headphones are not active; G _{_nx} is the transfer function from external noise to the feedforward microphone; and G _{_ffe} is the transfer function of the filtered signal through the output transducer to the ear when the feedback active noise cancellation signal path is active. The predetermined target response is a target transparency insertion gain ( _Thtig ), which is 0 dB in the transparency passband and is the same as the insertion gain achieved by the first feedforward filter outside the transparency passband.