Hearing protection and situational awareness system
A hearing protection and situational awareness system includes a wearable device, speakers, one or more beamformers, a microphone array, and a computation unit. The system generates a three-dimensional (3D) binaural sound for enhanced situational awareness; provides hearing protection by active noise cancelation; provides hearing enhancement by automatic gain control; and performs background noise reduction and cancelation. The system performs automated sound detection, identification, and localization, with automated voice assistance, and facilitates clear two-way communications. Each beamformer(s) outputs a sound track associated with a sound captured by the microphone array in a direction(s) of an acoustic beam pattern(s). The computation unit combines filtered sound tracks generated using head-related transfer function (HRTF) filters into left and right sound channels to drive the speaker(s) in left and right hearing members of the wearable device, respectively, thereby generating a 3D binaural sound including cues of the sound source directions.
1. A system for providing hearing protection and situational awareness, the system comprising:
a wearable device comprising hearing members, wherein the hearing members comprise a left hearing member configured to be disposed on a left ear of a user, and a right hearing member configured to be disposed on a right ear of the user;
one or more speakers disposed at an inner location of each of the hearing members of the wearable device;
a sound source localization unit operably coupled to the left hearing member and the right hearing member of the wearable device, the sound source localization unit comprising:
a plurality of beamformers, each configured to generate one acoustic beam pattern pointing to one direction;
an array of four or more outer microphones comprising two or more microphones disposed at outer locations of each of the left hearing member and the right hearing member of the wearable device, wherein the array of four or more outer microphones is configured to capture sound from the plurality of acoustic beam patterns, pointing to outside of the hearing members, and wherein each of the plurality of beamformers is configured to output a sound track, each sound track associated with the captured sound in the direction of the corresponding acoustic beam pattern;
a pair of inner microphones comprising one inner microphone disposed at an inner location of each of the left hearing member and the right hearing member of the wearable device, wherein the pair of inner microphones is configured to facilitate active noise cancelation;
a computation unit disposed in one or more of the hearing members of the wearable device and operably coupled to the array of four or more outer microphones and the pair of inner microphones disposed at an inner location of each of the hearing members of the wearable device, the computation unit comprising:
at least one processor;
a memory unit operably and communicatively coupled to the at least one processor, wherein the memory unit is configured to store a database of head-related transfer function filters, and computer program instructions defined by a plurality of signal processing modules and executable by the at least one processor;
the signal processing modules comprising at least one sound generator configured to:
for each of the plurality of beamformers, retrieve a pair of head-related transfer function filters associated with the direction of the plurality of acoustic beam patterns generated by each of the plurality of beamformers, from the database;
apply the retrieved corresponding pair of head-related transfer function filters to the output sound track from each of the plurality of beamformers, to generate a pair of filtered sound tracks, one for the left ear and another one for the right ear, wherein each of the filtered sound tracks represents the sound captured from the direction of the corresponding acoustic beam pattern;
combine the filtered sound tracks generated for the left ear into a left sound channel to drive at least one of the one or more speakers in the left hearing member of the wearable device, and combine the filtered sound tracks generated for the right ear into a right sound channel to drive at least one of the one or more speakers in the right hearing member of the wearable device, thereby generating a binaural sound comprising cues of the sound source directions, wherein the one or more speakers in the each of the hearing members of the wearable device reproduce real sound outside of the wearable device, whereby output sounds from the wearable device comprise a cue of at least one of one or more sound source directions;
the signal processing modules further comprising an artificial intelligence-enabled sound identification module configured to automatically recognize characteristics of the captured sound and identify categories of the captured sound using one or more pretrained artificial intelligence models;
the signal processing modules further comprising a voice assistant operably coupled to the one or more speakers and, in communication with the artificial intelligence-enabled sound identification module and the sound source localization unit, configured to determine one or more sound source locations from the output sound track of each of the plurality of beamformers and to report the identified categories of the captured sound and a direction of the captured sound with voice assistance via one of the one or more speakers and one or more other interface devices; and
the signal processing modules further comprising a set of active noise cancellation units and a set of noise reduction units for utilizing information in both a spatial domain and a temporal domain to cancel and/or reduce noise, thereby ensuring hearing protection and improving signal-to-noise ratio.
2. The system of claim 1 , wherein the pair of inner microphones is configured as a noise dosimeter to monitor and record sound levels within the hearing members of the wearable device, and to communicate the recorded sound levels to a storage unit for noise level analysis and review activities.
3. The system of claim 1 , wherein the signal processing modules further comprise an automatic gain control unit configured to control a level of output sound of the one or more speakers by selectively increasing gain when a level of the output sound is substantially low for hearing enhancement and situational awareness, and reducing the gain when the level of the output sound is substantially high for hearing protection.
4. The system of claim 1 , further comprising a supplementary microphone operably coupled to an adjustable arm extending from one of the hearing members of the wearable device, wherein the supplementary microphone is configured to allow the user to speak thereinto for facilitating two-way communications.
5. The system of claim 1 , wherein the wearable device is a headset, and wherein the hearing members are ear cups.
6. The system of claim 1 , wherein the set of active noise cancellation units use sound from each of other of the plurality of beamformers to cancel the noise in each of the plurality of beamformers, thereby using spatial information to perform noise cancelation, and thereby improving the signal-to-noise (SNR) of the output of the beamformers.
7. The system of claim 1 , wherein, for each sound track, the set of active noise cancellation units apply models or filters to reduce background noise, thereby using temporal information for noise reduction.
8. The system of claim 1 , wherein the set of active noise cancellation units execute active noise cancelation to cancel background noise that leaks into the hearing members.
9. A system for providing hearing protection and situational awareness, the system comprising:
a wearable device comprising hearing members, wherein the hearing members comprise a left hearing member configured to be disposed on a left ear of a user, and a right hearing member configured to be disposed on a right ear of the user;
one or more speakers disposed at an inner location of each of the hearing members of the wearable device;
a sound source localization unit operably coupled to the left hearing member and the right hearing member of the wearable device, the sound source localization unit comprising:
a plurality of beamformers, each configured to generate one acoustic beam pattern pointing to one direction;
an array of four or more outer microphones comprising two or more microphones disposed at outer locations of each of the left hearing member and the right hearing member of the wearable device, wherein the array of four or more outer microphones is configured to capture sound from the plurality of acoustic beam patterns, pointing to outside of the hearing members, and wherein each of the plurality of beamformers is configured to output a sound track, each sound track associated with the captured sound in the direction of the corresponding acoustic beam pattern;
a pair of inner microphones comprising one inner microphone disposed at an inner location of each of the left hearing member and the right hearing member of the wearable device, wherein the pair of inner microphones is configured to facilitate active noise cancelation;
a computation unit disposed in one or more of the hearing members of the wearable device and operably coupled to the array of four or more outer microphones and the pair of inner microphones disposed at an inner location of each of the hearing members of the wearable device, the computation unit comprising:
at least one processor;
a memory unit operably and communicatively coupled to the at least one processor, wherein the memory unit is configured to store a database of head-related transfer function filters, and computer program instructions defined by a plurality of signal processing modules and executable by the at least one processor; and
the signal processing modules comprising:
at least one sound generator configured to:
for each of the plurality of beamformers, retrieve a pair of head-related transfer function filters associated with the direction of the plurality of acoustic beam patterns generated by each of the plurality of beamformers, from the database;
apply the retrieved corresponding pair of head-related transfer function filters to the output sound track from each of the plurality of beamformers, to generate a pair of filtered sound tracks, one for the left ear and another one for the right ear, wherein each of the filtered sound tracks represents the sound captured from the direction of the corresponding acoustic beam pattern;
combine the filtered sound tracks generated for the left ear into a left sound channel to drive at least one of the one or more speakers in the left hearing member of the wearable device, and combine the filtered sound tracks generated for the right ear into a right sound channel to drive at least one of the one or more speakers in the right hearing member of the wearable device, thereby generating a binaural sound comprising cues of the sound source directions, wherein the one or more speakers in the each of the hearing members of the wearable device reproduce real sound outside of the wearable device, whereby output sounds from the wearable device comprise a cue of at least one of one or more sound source directions;
the signal processing modules further comprising an artificial intelligence-enabled sound identification module configured to automatically recognize characteristics of the captured sound and identify categories of the captured sound using one or more pretrained artificial intelligence models;
the signal processing modules further comprising a voice assistant operably coupled to the one or more speakers and, in communication with the artificial intelligence-enabled sound identification module and the sound source localization unit, configured to determine one or more sound source locations from the output sound track of each of the plurality of beamformers and to report the identified categories of the captured sound and a direction of the captured sound with voice assistance via one of the one or more speakers and one or more other interface devices;
the signal processing modules further comprising a set of active noise cancellation units and a set of noise reduction units for utilizing information in both a spatial domain and a temporal domain to cancel and/or reduce noise, thereby ensuring hearing protection and improving signal-to-noise ratio; and
the signal processing modules further comprising an automatic gain control unit configured to control a level of output sound of the one or more speakers by selectively increasing gain when a level of the output sound is substantially low for hearing enhancement and situational awareness, and reducing the gain when the level of the output sound is substantially high for hearing protection.
10. The system of claim 9 , wherein the pair of inner microphones is configured as a noise dosimeter to monitor and record sound levels within the hearing members of the wearable device, and to communicate the recorded sound levels to a storage unit for noise level analysis and review activities.
11. The system of claim 9 , further comprising a supplementary microphone operably coupled to an adjustable arm extending from one of the hearing members of the wearable device, wherein the supplementary microphone is configured to allow the user to speak thereinto for facilitating two-way communications.
12. The system of claim 9 , wherein the wearable device is a headset, and wherein the hearing members are ear cups.
13. The system of claim 9 , wherein the set of active noise cancellation units use sound from each of other of the plurality of beamformers to cancel the noise in each of the plurality of beamformers, thereby using spatial information to perform noise cancelation, and thereby improving the signal-to-noise (SNR) of the output of the beamformers.
14. The system of claim 9 , wherein, for each sound track, the set of active noise cancellation units apply models or filters to reduce background noise, thereby using temporal information for noise reduction.
15. The system of claim 9 , wherein the set of active noise cancellation units execute active noise cancelation to cancel background noise that leaks into the hearing members.