Wearable audio output apparatus, system, and method for delivering user-specific audio output to user

Abstract

A method (400) for delivering user-specific audio output to a user is disclosed. The method (400) includes receiving audio input from a user device (106) associated with the user. Further, the method (400) includes identifying auditory stimulus from the audio input. The auditory stimulus being indicative of one or more biomarkers that trigger a neural response from the brain of the user. Furthermore, the method (400) includes obtaining, from the one or more biosignal electrodes (104), electroencephalogram (EEG)-based neural response data corresponding to the auditory stimulus. Moreover, the method (400) includes identifying a frequency range associated with altered response patterns based on the (EEG)-based neural response data. The method (400) further includes modulating amplitude corresponding to the frequency range. Finally, the method (400) includes delivering, via a speaker unit (214) of a wearable audio output apparatus (102), a user-specific audio output to the user based on the modulation.

Description

WEARABLE AUDIO OUTPUT APPARATUS, SYSTEM, AND METHOD FOR DELIVERING USER-SPECIFIC AUDIO OUTPUT TO USERFIELD OF THE INVENTION

[0001] The present disclosure, in general, relates to the field of audio signal processing. More particularly, the present invention relates to a wearable audio output apparatus, a system, and a method for delivering user-specific audio output to a user.BACKGROUND

[0002] Conventional methods for assessing hearing capability primarily focus on the health of the eardrum and the auditory pathway up to the tympanic membrane. However, these approaches are not designed to detect sensorineural impairments or deficits within the neural pathways of the auditory- system.[0003 Audiometry is among the most widely used clinical procedures for evaluating an individual's hearing competence. It determines the severity of hearing loss and identifies which parts of the auditory system (outer, middle, or inner ear), may be affected. Typically, the audiometry employs an audiometer that generates sounds at varying frequencies and volumes to assess auditory perception.

[0004] A commonly practiced technique within audiometry is Pure Tone Audiometry, which involves presenting tones at different frequencies and intensities, requiring the patient to signal, usually by pressing a button, upon hearing a tone. While widely adopted, this method relies heavily' on the patient’s subjective response, which can compromise accuracy. Factors such as cooperation, attention, cognitive ability, and even malingering introduce variability in test outcomes. This dependency poses challenges for individuals such as young children, elderly patients with cognitive impairments, or those with limited mobility, who may struggle to respond consistently. Consequently, the reliability of results is closely tied to the patient’s ability- to provide accurate feedback.

[0005] Another traditional method, tympanometry, evaluates middle ear function by measuring changes in ear canal pressure, offering insights into the health of the eardrum and middle ear structures. However, this technique does not measure sensorineural hearing loss (SNHL), which is often associated with damage to the inner ear or auditory nerve. Furthermore, it fails to provide information on the functionality of the auditory cortex or neural pathways responsible for soundprocessing. As a result, conventional hearing tests frequently fall short in detecting early-stage SNHL and auditory pathway deficits arising from neurological conditions or aging.

[0006] These traditional assessments generally rely on manual methods requiring active patient participation. Such dependence on user intervention can lead to inaccurate or inconsistent results due to variations in attentiveness or response reliability.

[0007] Conventional audio systems including, earphones, speakers, and hearing devices, are typically designed to deliver sound output that is uniform across users. These systems do not account for variation in individual auditory perception or neural responses, as the same audio output may be perceived differently’ by different users due to their hearing capabilities.

[0008] Further, existing adaptive audio technologies often rely on environmental sensors to detect the sound. However, such approaches lack direct insight into how a user’s brain actually react to the audio output. Therefore, these existing technologies cannot dynamically optimise the audio output to align with the user's real-time neurological state.

[0009] Therefore, there exists a need for a system and / or method that may overcome that addresses the limitations of at least the above conventional methods.SUMMARY

[0010] This summary is provided to introduce a selection of concepts, in a simplified format, that are further described in the detailed description of the disclosure. This summary is neither intended to identify key or essential inventive concepts of the disclosure nor is it intended to determine the scope of the disclosure.

[0011] In an aspect of the present disclosure, a wearable audio output apparatus for a user is disclosed. The wearable audio output apparatus includes a body adapted to be worn proximal to at least one ear of the user. The body includes one or more biosignal electrodes coupled thereto. The wearable audio output apparatus further includes a processor in communication with the one or more biosignal electrodes. The processor is configured to receive audio input from a user device associated with the user. T he processor is configured to identify auditory stimulus from the audio input. The auditory’ stimulus being indicative of one or more biomarkers that trigger a neural response from the brain of the user. Further, the processor is configured to obtain, from the one or more biosignal electrodes, electroencephalogram (EEG)-based neural response data corresponding to the auditory stimulus. Furthermore, the processor is configured to identify a frequency range associated with altered response patterns based on the (EEG)-based neural response data. Moreover, the processor is configured to modulate amplitude corresponding to the frequency range. Finally, the processor isconfigured to deliver, via a speaker unit, a user-specific audio output to the user based on the modulation.

[0012] In another aspect of the present disclosure, a method for delivering user-specific audio output to a user is disclosed. The method includes receiving audio input from a user device associated with the user. Further, the method includes identifying auditory stimulus from the audio input. The auditory stimulus being indicative of one or more biomarkers that trigger a neural response from the brain of the user. Furthermore, the method includes obtaining, from the one or more biosignal electrodes, electroencephalogram (EEG)-based neural response data corresponding to the auditory stimulus. Moreover, the method includes identifying a frequency range associated with altered response patterns based on the (EEG)-based neural response data. The method further includes modulating amplitude corresponding to the frequency range. Finally, the method includes delivering, via a speaker unit of a wearable audio output apparatus, a user-specific audio output to the user based on the modulation.

[0013] In yet another aspect of the present disclosure, a system for delivering user-specific audio output to a user is disclosed. The system includes a wearable audio output apparatus worn by the user, proximal to at least one ear of the user. The system further includes one or more biosignal electrodes coupled to the wearable audio output apparatus. Furthermore, the system includes a processor in communication with the one or more biosignal electrodes and the wearable audio output apparatus. The processor is configured to receive audio input from a user device associated with the user. The processor is configured to identify auditory stimulus from the audio input. The auditory stimulus being indicative of one or more biomarkers that trigger a neural response from the brain of the user. Further, the processor is configured to obtain, from the one or more biosignal electrodes, electroencephalogram (EEG)-based neural response data corresponding to the auditory stimulus. Furthermore, the processor is configured to identify a frequency range associated with altered response patterns based on the (EEG)-based neural response data. Moreover, the processor is configured to modulate amplitude corresponding to the frequency range. Finally, the processor is configured to deliver, via a speaker unit, a user-specific audio output to the user based on the modulation.

[0014] To further clarify the advantages and features of the present invention, a more particular description of the invention will be rendered by reference to specific embodiments thereof, which are illustrated in the appended drawings. It is appreciated that these drawings depict only typical embodiments of the invention and are therefore not to be considered limiting of its scope. Theoinvention will be described and explained with additional specificity and detail in the accompanying drawings.BRIEF DESCRIPTION OF THE DRAWINGS

[0015] These and other features, aspects, and advantages of the present invention will become better understood when the following detailed description is read with reference to the accompanying drawings in which like characters represent like parts throughout the drawings, wherein:

[0016] Figure 1 illustrates an environment for implementing a system for delivering user-specific audio output to a user, in accordance with an embodiment of the present disclosure:

[0017] Figure 2A illustrates the wearable audio output apparatus in the form of a headband, in accordance with an embodiment of the present disclosure;

[0018] Figure 2B illustrates the wearable audio output apparatus in the form of an earphone, in accordance with an embodiment of the present disclosure;

[0019] Figure 2C illustrates the wearable audio output apparatus in the form of a smart glasses, in accordance with an embodiment of the present disclosure;

[0020] Figure 3 illustrates a block diagram of the system for delivering the user-specific audio output to the user, in accordance with an embodiment of the present disclosure;

[0021] Figure 4 illustrates a flowchart depicting a method for delivering the user-specific audio output to the user, in accordance with an embodiment of the present disclosure; and

[0022] Figure 5 illustrates a flowchart depicting an exemplary method for real-time tracking and diagnosis of hearing loss using biosignals, in accordance with an embodiment of the present disclosure.

[0023] Further, skilled artisans will appreciate that elements in the drawings are illustrated for simplicity and may not have been drawn to scale. For example, the flow charts illustrate the method in terms of the most prominent operations involved to help improve understanding of aspects of the present inventive concepts. Furthermore, in terms of the construction of the device, one or more components of the device may have been represented in the drawings by conventional symbols, and the drawings may show only those specific details that are pertinent to understanding the example embodiments of the present inventive concepts so as not to obscure the drawings with details that will be readily apparent to those of ordinary skill in the art having the benefit of the description herein.DETAILED DESCRIPTION OF FIGURES

[0024] For the purpose of promoting an understanding of the principles of the inventive concepts, reference will now be made to example embodiments illustrated in the drawings and specific language will be used to describe the same. It will nevertheless be understood that no limitation of the scope of the inventive concepts is thereby intended, such alterations and further modifications in the illustrated system, and such further applications of the principles of the inventive concepts as illustrated therein being contemplated as would normally occur to one skilled in the art to which the inventive concepts relate.

[0025] It will be understood by those skilled in the art that the foregoing general description and the following detailed description are explanatory of the present disclosure and are not intended to be restrictive thereof.

[0026] Reference throughout this specification to “an aspect”, “another aspect” or similar language means that a particular feature, structure, or characteristic described in connection with the example embodiments is included in at least one example embodiment of the present disclosure. Thus, appearances of the phrase “in example embodiments”, “in another example embodiment” and similar language throughout this specification may, but do not necessarily, all refer to the same example embodiments.

[0027] The terms "comprises", "comprising", or any other variations thereof, are intended to cover a non-exclusive inclusion, such that a process or method that comprises a list of operations does not include only those operations but may include other operations not expressly listed or inherent to such process or method. Similarly, one or more devices or sub-systems or elements or structures or components proceeded by "comprises... a" does not, without more constraints, preclude the existence of other devices or other sub-systems or other elements or other structures or other components or additional devices or additional subsystems or additional elements or additional structures or additional components.

[0028] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skilled in the art to which this disclosure belongs. The system, methods, and examples provided herein are illustrative only and not intended to be limiting.

[0029] Figure 1 illustrates an environment for implementing a system 100 for delivering user- specific audio output to a user, in accordance with an embodiment of the present disclosure. Theuser-specific audio output herein to audio signal produced based on hearing capabilities of a particular user. In an embodiment, the environment 1000 may include a system 100 that may include a wearable audio output apparatus 102 worn by the user, proximal to at least one ear of the user. In an embodiment, the wearable audio output apparatus 102. The wearable audio output apparatus 102 may interchangeably be termed as the apparatus 102 or the wearable audio output device 102. or the device 102 within the scope of the present disclosure.

[0030] The wearable audio output apparatus 102 may include but are not limited to earphone, headphone, earbuds, smart glasses, headband, cap, earplugs, extended reality (XR) Headsets, bone conduction audio devices depending on the user's comfort or any other apparatus 102 which may be wear proximal to at least one ear of the user or a cephalic region of the user and capable of delivering audio output to the user.

[0031] Figure 2A illustrates the wearable audio output apparatus 102 in the form of the headband, in accordance with an embodiment of the present disclosure. In an embodiment, the headbands may be adapted to be worn proximal to the cephalic region of the user. The headbands may be integrated with a speaker unit 214. Further, the headband may also be integrated with a microphone 218. Figure 2B illustrates the wearable audio output apparatus 102 in the form of the earphone, in accordance with an embodiment of the present disclosure. For example, the wearable audio output apparatus 102 is a brainwave mapping earphone designed to fit within an ear canal of the user. The earphone may include the speaker unit 214 and tire biosignal electrodes 104 that may be disposed on a connecting arm 208. The connecting arm 208 may be mechanically supported by a dynamic size -adjusting mechanism 2.10. The dynamic size-adjusting mechanism 304 may be configured to adapt the positioning of the biosignal electrodes 104 in proximity to the ear or the cephalic region of the user. Figure 2C illustrates the wearable audio output apparatus 102 in the form of the smart glasses, in accordance with an embodiment of the present disclosure. As shown, the smart glass may be adapted to be worn by the user. The smart glasses may include an electrode array or pad 212 that may include the biosignal electrodes 104 positioned to establish contact with the user’s skin, particularly in proximity to the cephalic region. The smart glass further includes a processor 302, an input interface (touch) unit 216. Further, the smart glasses may include the speaker unit 2.14 and the microphone 2.18.

[0032] Again, referring to Figure 1, the environment 1000 may include a user device 106 on which an audio may be played. The user device 106 may be wired or wirelessly connected to the wearable audio output apparatus 102.

[0033] Further, the wearable audio output apparatus 102 may include a body that may be wore by the user. The wearable audio output apparatus 102 may- further include one or more biosignal electrodes 104 that may be coupled to the body. Further, the wearable audio output apparatus 102 may include the speaker unit 214 disposed within the body. In an embodiment, the one or more biosignal electrodes 104 may herein refer to electroencephalogram (EEG) electrodes 104 or EEG sensors 104 or the biosignal electrodes 104 within the scope of the present disclosure. The biosignal electrodes 104 may be adapted to maximize contact while minimizing discomfort, leveraging advanced noise-cancellation algorithms to improve signal fidelity in real-world conditions. These biosignal electrodes 104 may be positioned proximal to a scalp, mastoid or concha region, based on the type of the wearable audio output apparatus 102. In an exemplary implementation, the placement of the EEG sensors may be crucial for capturing responses from both auditory cortices. The EEG sensors may be positioned over the temporal lobes or within the ear canal, particularly when using earphones for brainwave mapping, to effectively monitor responses from both the left and right hemispheres.

[0034] Figure 3 illustrates a block diagram of the system 100 for delivering the user-specific audio output to the user, in accordance with an embodiment of the present disclosure.

[0035] In an embodiment, the system 100 may include the at least one processor 302 (also referred to as the processor 302 integrated in the wearable audio output apparatus 102), a memory 304, a plurality of modules 306, and the wearable audio output apparatus 102. " Die at least one processor 302, the memory 304, the plurality of modules 306, and the wearable audio output apparatus 102 are communicably coupled with each other.

[0036] In an embodiment, the at least one processor 302 may be in communication with the memory 304 and the biosignal electrodes 104. The at least one processor 302 may be a single processing unit or several units, all of which could include multiple computing units. The at least one processor 302 may be implemented as one or more microprocessors, microcomputers, microcontrollers, digital signal processors, central processing units, state machines, logic circuitries, and / or any devices that manipulate signals based on operational instructions. Among other capabilities, the at least one processor 302 may be configured to fetch and execute computer-readable instructions and data stored in the memory 304.

[0037] In an embodiment, the memory 304 may include any non-transitory computer-readable medium known in the art including, for example, volatile memory, such as static random access memory (SRAM) and dynamic random access memory (DRAM), and / or non-volatile memory, suchas read-only memory (ROM), erasable programmable ROM, flash memories, hard disks, optical disks, and magnetic tapes.

[0038] In an embodiment, the plurality of modules 306 may be configured to deliver the user- specific audio output to the user.

[0039] In one embodiment, the plurality of modules 306 may include the one or more instructions (stored in a memory 304) that may be executed to cause the system 100, in particular, the processor 302 of the system 100, to perform the one or more functions / methods, as discussed here in the present disclosure. In one embodiment, the plurality of modules 306 may be implemented at least in part as hardware, which may work in conjunction with the instructions to perform the functions / methods discussed herein. In an embodiment, the plurality of modules 306 may be implemented using one or more artificial intelligence (Al) units or foundational models that may include a plurality of neural network layers. Examples of neural networks include, but are not limited to, transformer neural network (TNN), convolutional neural network (CNN), deep neural network (DNN), recurrent neural network (RNN), and Restricted Boltzmann Machine (RBM). Further, "learning’ may be referred to in the disclosure as a method for training a predetermined target device (for example, a robot) using a plurality of learning data to cause, allow, or control the target device to decide or predict. Examples of learning techniques include, but are not limited to, supervised learning, unsupervised learning, semi -supervised learning, reinforcement learning, or self-attention mechanisms. At least one of a plurality of TNN, CNN, DNN, RNN, RMB models and the like may be implemented to thereby achieve execution of the present subject matter’s mechanism through an Al model. A function associated with an Al unit may be performed through the non-volatile memory, the volatile memory, and the processor. The processor 302 may include one or a plurality of processors. At this time, one or a plurality of processors may be a general-purpose processor, such as a central processing unit (CPU), an application processor (AP), or the like, a graphics-only processing unit such as a graphics processing unit (GPU), a visual processing unit (VPU), an optical neural network (ONN) and / or an Al-dedicated processor, such as a neural processing unit (NPU). One or a plurality of processors control the processing of the input data in accordance with a predefined operating rale or artificial intelligence (Al) model stored in the non-volatile memory and the volatile memory'. The predefined operating rule or artificial intelligence model is provided through training or learning.

[0040] The plurality of modules 306 may' include a receiving module 322, an identifying module 324, an obtaining module 326, a modulating module 328, and a delivering module 330. In an embodiment, the receiving module 322, the identifying module 324, the obtaining module 326, themodulating module 328, and the delivering module 330 may be in communication with each other. The working of the plurality of modules 306 may be explained in conjunction with Figure 4. 0041] Figure 4 illustrates a flowchart depicting a method 400 for delivering the user-specific audio output to the user, in accordance with an embodiment of the present disclosure. The method 400 is a computer-implemented method 400 whose steps may be performed by the wearable audio output apparatus 102, more particularly, the processor of the apparatus 102. In another embodiment, a cloud server may perform the steps. At step 402, the method 400 may include receiving, via the receiving module 322, audio input from the user device 106 associated with the user. In an exemplary’ implementation, the apparatus 102 may be worn in the ear, such that the biosignal electrodes 104 may establish contact with ear canal of the ear. In an exemplary embodiment, when the audio input may be played on the user device, the processor 302. may receive the audio input. In a non-limiting example, the audio input may be a song played on the user device 106, the audio generated from a video played on the user device 106, or any other media played on the user device 106. In another example, the audio input may be audio received from another user device during on-going call.

[0042] At step 404, the method 400 may include identifying, via the identifying module 324, auditory stimulus from the audio input. In an embodiment, the auditory' stimulus being indicative of one or more biomarkers that trigger a neural response from the brain of the user. In particular, when the auditory stimulus may be given to the user, the biosignal electrodes 104 may utilized the EEG electrodes 104 to track auditory’ responses, also refer to as EEG-based neural responses within the scope of the present disclosure. The EEG-based neural responses may offer deeper insights into sensorineural health that traditional methods may not capture. Similarly, EEG-based headband electrodes (the biosignal electrodes 104) may be worn over the scalp or forehead to establish contact w'ith a specific body part for real-time capturing of the biosignals.

[0043] Further, at step 406, the method 400 may include obtaining, via the obtaining module 326 from the biosignal electrodes 104, EEG-based neural response data corresponding to the auditory stimulus provided to the user. The EEG-based neural response data may herein refer to EEG-based responses or EEG responses or neural responses or auditory responses, within the scope of the present disclosure. In an exemplary embodiment, the processor may prioritize capturing or obtain the EEG¬ based neural response data at least within a specific short time window which may be considered as a critical diagnostic for neural -based hearing impairments.

[0044] In an implementation, the processor 302 may calibrate frequency-specific auditory' stimulus to evoke that the EEG-based neural responses that may be captured and analyzed through EEGresponse captured through the biosignal electrodes 104. Such responses, termed herein and after as Audio Stimulus Response (ASR), may- provide objective data on the brain’s reaction to the audio.

[0045] In an embodiment, the processor may process the EEG-based auditory responses (also referred to as the EEG-based neural response data) after being calibrated. The processing of the EEG¬ based auditory responses may provide a direct, objective measure of neural activity in response to the auditory stimulus, which may bypass the need a subjective input.

[0046] In an exemplary scenario, when the audio stimulus is presented, an auditory pathway may process the audio through the outer, middle, and inner ear, eventually transmitting the signal to an auditory cortex. At this stage. Auditory- Evoked Potentials (AEPs) may be detected via the EEG. These AEPs may be electrophysiological responses generated by the brain’s processing of the a, allowing for a real-time assessment of auditory neural function. Since the EEG responses from the biosignal electrodes 104 may correlate precisely with specific characteristics of the auditory stimulus, such as frequency, amplitude, and phase, therefore, hearing thresholds and sensory processing in the auditory cortex may be objectively measured. For example, sustained responses such as Steady-State Auditory Evoked Potentials (SSAEPs) may be triggered by repetitive auditory stimulus at particular frequencies, directly linking neural activity to the input frequency. Such EEG responses may serve as a robust indicator of auditory function, helping map neural responses to different sound frequencies and intensities without patient involvement. By analyzing these responses, EEG-based hearing tests may objectively determine hearing acuity and identify thresholds for each frequency in the auditory range.

[0047] In another embodiment, the processed EEG-based neural response data may be recorded in response to calibrated auditory stimulus to effectively detect sensorineural hearing loss.

[0048] In an exemplary scenario, the processing of the EEG-based neural response data may¬ leverage encoding the auditory stimulus for accessibility. The system 100 may translate auditory test signals into accessible formats like familiar tones, music, or media. The system 100 may allow users to track and interpret their auditory health in an intuitive way. Further, the system 100 may employ Digital signal processing (DSP) convert complex auditory responses into recognizable sounds, which may enable the users to associate specific tones or melodies with their hearing status.

[0049] In an embodiment, the processor 302 may employ the ML model or the model to continuously- analyze hearing patterns (also referred to as response paterns) from the EEG-based neural response data to enhance diagnostic precision over time. ’lire hearing patterns or the response patterns may involve various factors such as pitch / frequency, intensity / loudness, clarity / distortion,and the ability to perceive speech in noisy environments. The hearing patterns may be altered due to factors like damaged hair cells in the cochlea, neural damage, or other auditory system malfunctions. Impairments may vary in type, for example, sensorineural, conductive, mixed, and severity such as mild, moderate, severe, or profound. The classification of hearing patterns may be further trained by applying the ML model using training data. The hearing patterns may be classified through the definition of rules and / or training through the application of machine learning to perform the classification of a certain degree of impairment. The rale-based framework may include predefined rules or heuristics that may be applied to simulate or analyze the hearing patterns or the response patterns. These rules may be based on clinical knowledge, for example, how certain frequencies are affected by specific types of hearing loss or standard audiological data, for example, audiogram curves, frequency response, speech intelligibility, or mathematical models that may simulate how the audio is processed by the system 100.

[0050] In an embodiment, the processor 302 may be configured to detect alerted response patterns based on the analyzing of the response patterns of the hearing patterns.

[0051] Further, at step 408, the method 400 may include identifying, via the identifying module 224, a frequency range associated with the alerted response patterns based on the (EEG)-based neural response data. Moreover, at step 410, the method 400 may include modulating, via the modulating module 328, amplitude corresponding to the frequency range.

[0052] In an exemplary embodiment, the processor 302 may employ an adaptive frequency-specific hearing compensation process that may use a real-time audio equalizer model that dynamically adjust or modulate the amplitude across a broad frequency spectrum. In an embodiment, the process initiates with a playback of a testing audio signal, constructed as spherical audio containing multiple discrete frequencies spanning the full range of human hearing. This spherical / spatial audio signal is designed to probe the user’s response across all key frequency regions individually, enabling a comprehensive, frequency-by-frequency assessment of auditory sensitivity. The EEG sensors (the biosignal electrodes 104) monitor neural activity in response to each frequency, capturing the neural responses in real-time to identify any diminished neural engagement in specific frequency bands, i.e. the frequency range associated with the alerted response patterns. The identification of the frequency range with the alerted response pattens provides a direct measure of auditory sensitivity and reveals frequency-specific deficiencies that conventional audio testing methods may overlook.

[0053] In an exemplary embodiment, the processor 302 may analyze the EEG-based neural response data to each frequency corresponding to the auditory stimulus. Further, the processor 302 may use a specialized signal processing and ML algorithms to detect reduced neural response amplitudes andthe alerted wavefonns (alerted response patterns) in the identified frequency range where the user exhibits diminished sensitivity. Further, the processor 302 may use a neuroequalizer module (the adaptive equalizer algorithm) to generate a personalized frequency response profile corresponding to the user. In an embodiment, for the identified frequency range with reduced neural activation, indicating possible hearing deficiencies, the processor 302 may be configured to modulate or adjust, for example increasing the amplitude or gain corresponding to the identified frequency range within the personalized frequency response profile, thereby enabling adjustment of the audio output in real-time. In particular, the processor 302 may adjust tone in real time to restore perceptual balance for under-responsive bands while preserving fidelity in unaffected frequency bands or range. In one configuration, the neuroequalizer operates as a prescription module that leverages Auditory Evoked Potentials, including Auditory Steady-State Responses (ASSR), to derive clinically relevant neuro-audiometric thresholds which are stored as a digital hearing prescription within the earphone (the wearable audio output apparatus 102).

[0054] In an exemplary embodiment, the processor 302 may analyze the EEG-based neural response data to each frequency. Further, the processor 302 may use a specialized algorithm to detect low neural response amplitudes in the identified frequency range where the user exhibits reduced sensitivity. Further, the adaptive equalizer algorithm to processes these findings to generate a personalized frequency response profile corresponding to the user.

[0055] In an embodiment, for the identified frequency range with reduced neural activation, indicating possible hearing deficiencies, the processor 302 may be configured to modulate or adjust, for example increasing the amplitude or gain corresponding to the identified frequency range within the personalized frequency response profile, thereby enabling adjustment of the audio output in real-time. In an advantageous effect, the adjustment of the audio output ensures that sounds in the identified frequency range are amplified to levels perceptible to the user. This frequency-targeted equalization approach acts as a real-time audio equalizer that tailors the auditory experience based on neural response data, thereby compensating for frequency-specific hearing impairments without disturbing the fidelity or balance of unaffected frequency bands.

[0056] In an embodiment, the system 100 may employ a baseline calibration phase in which reference EEG-based neural response data that is recorded across frequency bands or range to establish a per-user baseline profile. In an advantageous effect, the baseline measures reduce false positives and enable the system 100 to differentiate transient neural fluctuations from genuine frequency-specific deficits. The baseline profiles are periodically updated to reflect environmental conditions, electrode placement variations, or physiological changes. The apparatus 102 mayperform automated neurofeedback hearing tests during idle or low-activity periods, presenting controlled stimuli and using resultant EEG data to update internal equalization parameters so that the earphones (the apparatus 102) remain self-diagnosing and self-calibrating over time. These baseline references minimize the risk of false positives and ensure accurate detection of genuine auditory deficiencies.

[0057] In another embodiment, the processor 302 may be configured to output auditory nerve health in real-time based on the analyzing of the response patterns or the hearing patterns. The auditory’ nerve health may be used for clinical treatment or may be used for recommendation to the user.

[0058] Further, at step 412, the delivering module 330 may be configured to deliver, via the speaker unit 214, the user-specific audio output to the user based on the modulation.

[0059] In an embodiment, the processor 302 may be configured to update the personalized frequency response profile in real time based on variations in the EEG-based neural response data corresponding to the auditory stimulus. Further, the processor 302 may be configured to dynamically refining audio output characteristics including at least one of gain, tone, and frequency response, such that a synchronization is maintained with the user’s neural response (auditory response). Furthermore, the processor may be configured to selectively one of increasing or decreasing amplitudes of one or more frequency components to facilitate adaptive auditory’ conditioning.

[0060] In an embodiment, the processor 302 may use neuroequalizer as a continuous learning module or the ML model to identify gradual shifts in auditory perception arising from one or more factors using the ML model. The one or more factors include at least one of aging, environmental changes, or neural adaptation. In an exemplary scenario, the neuroequalizer monitors the EEG-neural response data during natural listening, updates a neural mapping model in real time, and incrementally fine-tunes compensation to track gradual perceptual shifts or also termed as the gradual shifts that arising from aging, environmental change, or neural adaptation. The equalizer supports taper positive and taper negative modes. In an embodiment, the taper positive modes progressively amplify frequencies exhibiting reduced neural engagement and the taper negative attenuates frequencies exhibiting excessive neural activation. These taper modes (taper positive and the taper negative) provide neurotherapeutic conditioning that reduces listening fatigue and promotes balanced neural activation over extended use.

[0061] In an embodiment, the processor 302 operates within a closed-loop adaptive framework. In an embodiment, the processor 302 is configured to obtain neurofeedback data corresponding to the one or more users from the user device 106 associated with each of the one or more users. Further,the processor 302 is configured to refine audio adjustment parameters based on the neurofeedback data. In another embodiment, the cloud server may be configured to obtain the neurofeedback data and refine audio adjustment parameters. Furthermore, the processor 302 is configured to detect, using the refined audio adjustment parameters, altered auditory response patterns from the EEG-based neural response data.

[0062] In an exemplary scenario, the user device 106 associated with each user locally records anonymized EEG-to-audio response pairs, which are periodically aggregated and transmitted to the cloud server without sharing raw data, thereby preserving privacy. The collective learning improves the accuracy of the ML model or the AI model for detecting hearing loss patterns and optimizing playback compensation. This closed-loop system evolves through population-scale neurofeedback, allowing the global model (the AI model or the ML model) to inform personalized adjustments while maintaining real-time responsiveness on the user device 106.

[0063] In an embodiment, the processor 302 may be configured to resolve audio ambiguity by leveraging crowdsourced neurofeedback data obtained through the federated learning.

[0064] In one embodiment, to resolve the audio ambiguity, the processor 302 may be configured to monitor the EEG-based neural response data while the audio is being played and to detect portions of the audio that produce weak, inconsistent, or otherwise ambiguous neural responses or inconsistent EEG engagement patterns, it marks those portions as targeted audio segments. The processor 302 further generates a compact ambiguity report for each segment. Each report includes only a segment identifier and a small set of numerical EEG-derived engagement scores, and is transmitted to the cloud server through the federated learning framework. Similar reports from many devices are aggregated by the cloud server to identify the targeted audio segments that consistently show poor neural engagement across a broad user population. The compact ambiguity report or the reports may herein to the crowdsourced neurofeedback data within the scope of the present disclosure.

[0065] In another embodiment, once the cloud server determines that a specific audio segment is ambiguous at a population level, it applies a transformer-based neural patching model to reconstruct that identified segment (the targeted audio segments) at the source. The model operates on a tokenized representation of the original audio, uses self-attention to analyze contextual structure within the segment, and predicts an improved token sequence that enhances clarity while preserving meaning. The refined tokens are decoded back into an audio waveform and seamlessly substituted into the original content file after boundary alignment and amplitude matching. Updated content containing the patched segment is then delivered to user devices, enabling the wearable apparatus toplay a neuro-tuned version of the audio that has been enhanced using crowdsourced EEG-derived feedback.

[0066] In an implementation, the processor 302 may leverage EEG microstate analysis for tracking neurobiological markers in patients (the users), potentially with Meniere’s Disease (MD), allowing for the differentiation between the MD and healthy controls (HC). In an embodiment, the processor 302, by identifying distinctive EEG microstate patterns associated with MD as well as other hearing disorders, the system 100 may provide a novel tool for tracking and managing challenges linked to the disease.

[0067] In an embodiment, the system 100 may include a dynamic audio output adaptation module that may be configured to assess auditory deficiencies in the patients (the users) with hearing disorders and correspondingly adapt the output audio signal to compensate for the hearing deficiencies in a such a way that the final audio perceived by the user is as same as the original audio signature of the file being played. In an implementation, the system 100 may leverage an array of modifiable auditory' stimulus parameters to elicit specific EEG responses, allowing for nuanced tracking of auditory health and cognitive functions. By adjusting these parameters systematically, the system 100 aims to capture robust neural responses that enhance the accuracy of diagnostics across conditions like hearing loss, cognitive impairment, and balance disorders such as Meniere’s Disease (MD).

[0068] In another implementation, the system 100 may leverage the frequency variation of the audio stimulus. The auditory stimulus at varying frequencies may be administered to the system 100 to test the full range of auditory neural responses. Frequency adjustments may allow the system 100 to examine how different sound frequencies activate distinct auditory pathways and how neural processing responds across the auditory frequency spectrum.

[0069] To minimize user burden and enable unobtrusive monitoring, the system 100 supports passive, timestamp-based stimulus extraction from natural audio content. The processor 302 tags acoustic events in ongoing playback, correlates EEG responses with these timestamped events, and incrementally refines the user’s hearing profile during normal listening. Aggregation of responses across diverse natural stimuli produces a statistically robust persona model that approaches clinical precision without discrete testing sessions, thereby enabling continuous long-term tracking of progressive hearing changes. Over time, as a listener (the user) is exposed to regular, unaltered audio such as music, speech, or ambient sounds, the system 100 continues to refine the user’s neural hearing profile in the background.

[0070] In another implementation, the system 100 may leverage the intensity or amplitude variation of the auditory stimulus. In an embodiment, by modifying the intensity or amplitude of the audio stimulus, the system 100 may assess neural sensitivity to sound (also referred to as the audio). For instance, lower-intensity sounds may reveal thresholds for auditory perception, while higher- intensity sounds may test neural tolerance and response latency. The intensity variations may also further help in evaluating compensatory mechanisms, as the patients with hearing loss may experience "loudness recruitment" or disproportionate perception of loud sounds, which may be accurately measured through the EEG data. The amplitude variation of stimulus audio may be particularly useful in identifying conditions where there may be a discrepancy in neural amplification, aiding in differentiating between neural and mechanical causes of hearing deficits.

[0071] In another implementation, the system 100 may leverage the phase variation of stimulus audio. The phase adjustments in auditory stimuli may offer insights into how the brain processes sound timing. By varying the phase, the system 100 may assess neural phase-locking capabilities, which are an important factor for auditory scene analysis, particularly in noisy environments, lire phase variation of stimulus audio may serve as a diagnostic tool for detecting temporal processing disorders and synchrony deficits in auditory' neural pathways, with implications tor assessing auditory temporal resolution in patients with cognitive impairment or early -stage neurodegenerative conditions.0072 In another implementation, the system 100 may leverage the directional variation of stimulus audio. The stimulus directionality may allow the system 100 to evaluate spatial hearing capabilities by introducing sound from different orientations. Such a setup may enable the assessment of binaural hearing and spatial localization, which may be crucial tor understanding a patient’s ability to detect sound sources in three-dimensional space. Further, patients with asymmetrical hearing loss or conditions such as Meniere’s Disease often display' impaired spatial orientation. Therefore, by using directional auditory stimuli, the system 100 may efficiently identify such spatial processing deficits and quantify compensatory efforts made by the brain. Additionally, the directional stimulus may aid in measuring neural asymmetry, where EEG signal discrepancies between the left and right hemispheres may provide valuable data for detecting unilateral neural hearing impairments.

[0073] In another implementation, the system 100 may incorporate spherical audio, providing a three-dimensional auditory’ experience by varying amplitude, frequency, and delay across multiple sound sources. The spherical audio, which may be simulated through methods like earphones placed on each side of the head, may leverage slight temporal delays, frequency modulations, and amplitude differences to mimic real-world sound perception. Such a three-dimensional audio simulation mayallow for a comprehensive evaluation of spatial hearing abilities, particularly aiding in the detection of hearing loss affecting sound localization and depth perception. By tracking how the auditory nerve may process these spatial cues, the system 100 may enable the identification of hearing loss characteristics unique to spherical audio perception.

[0074] In another implementation, the system 100 may incorporate audible range triangulation (ART). Under spherical audio, the ART may determine specific hearing loss within certain ranges. Using a combination of varying amplitudes, frequencies, and delays, the system 100 may triangulate the precise auditory ranges in which the patient may exhibit hearing loss. ML and artificial intelligence (Al) algorithms may assist in processing the EEG data by identifying patterns in neural responses to diverse spherical audio cues. By systematically adjusting these parameters, tire ART may pinpoint the hearing loss within specific auditory ranges, which may provide a high-resolution diagnostic profile of the patient’ s auditor}' deficits.

[0075] The ART Algorithm may further refine the triangulation process by integrating frequency- encoded audio and the ASR data to systematically determine the auditory ranges impacted by hearing loss. Through iterative analysis of the EEG responses across varied sound angles, amplitudes, and delays, the ART Algorithm may map the neural hearing response profile with high accuracy. Such an approach may automate the detection of hearing loss ranges and enable targeted adjustments to hearing aids or auditory interventions by providing a precise and individualized approach to managing sensorineural hearing impairment.

[0076] In another embodiment, the processor 302 may be configured to analyze the EEG-based neural response data for the diagnosis of hearing loss, which is discussed and explained in conjunction with Figure 5,

[0077] Figure 5 illustrates a flowchart depicting an exemplary method 500 for real-time tracking and diagnosis of hearing loss using biosignals, in accordance with an embodiment of the present disclosure.

[0078] At step 502, the biosignal electrodes 104 may capture one or more biosignals. The biosignals may be EEG biosignals. The processor 302, may utilize the EEG biosignals to obtain the EEG-based neural response data. The EEG-based neural response data is processed to measure essential parameters such as response latency, amplitude, and frequency of evoked potentials. Further, the processor 302 may be configured to identify variations in the parameters. The variations may indicate hearing impairment or neurological issues impacting auditory processing. For instance, prolonged latency in the EEG response may suggest issues with sound conduction or neural transmission.signalling potential SNHL. Additionally, changes in response amplitude may reveal reduced auditory sensitivity or cortical processing inefficiencies.

[0079] The EEG biosignals, for assessing the extent of hearing ability, may enable highly accurate detection of hearing deficits by providing an objective, non-invasive evaluation of auditory function, making the present method 400 and system 100 useful in situations where patient responses may be unreliable or unavailable. The EEG-based hearing tests may facilitate early intervention by identifying sensorineural deficits before they become clinically evident, offering a proactive approach to managing hearing health. Furthermore, the EEG responses may be integrated with machine learning (ML) algorithm(s) or model(s) that may allow continuous model improvement by enhancing predictive accuracy for hearing thresholds and ensuring robust tracking of auditory health over time.

[0080] At step 504, the processor 302 may be configured to monitor the auditory response. In an embodiment, the processor 302 may be configured to perform real-time ARM analysis that may enable prompt detection of auditory nerve degeneration, advancing early-stage intervention and management options.

[0081] At step 506, the processor 302 may be configured to perform microstate tracking. The microstate tracking may involve tracking of an EEG-based microstate whose characteristics may be captured, which may reveal compensatory' cognitive mechanisms, which may support balance and spatial orientation, but may also heighten subjective symptom perception in MD patients. These microstate features may also offer insights for personalizing cognitive compensation strategies and serve as potential diagnostic markers, facilitating early detection and intervention for Meniere's disease.

[0082] At step 508, the processor 302 may be configured to perform vertigo testing adaptation to assess the auditory '-vestibular interactions in the patients (the users) with vertigo or vestibular disorders.

[0083] In an embodiment of the present disclosure, the system 100 may render both-side static audio mapping techniques. In this case, the auditory stimulus is presented simultaneously to both ears. Such an approach may elicit bilateral auditory' responses, which may' be captured by the EEG sensors (the biosignal electrodes 104) as bilateral Auditory Evoked Potentials (AEPs). The auditory' stimulus may consist of calibrated tones, clicks, or frequency-modulated sounds across the audible frequency range, such as 250 Hz to 8,000 Hz. In an embodiment, delivering the auditory stimulus simultaneously to both ears may allow the assessment of synchronous auditory processing andprovide insight into how well the auditory pathways function together. By delivering the same auditory’ input to both ears, the present disclosure may' evaluate interaural symmetry in the neural response, helping to assess how similarly each ear's auditory pathway processes the signal.

[0084] In an implementation, bilateral hearing function may be assessed by presenting the auditory stimulus simultaneously to both ears. Such an approach may establish a baseline for evaluating bilateral hearing performance, which may be valuable for diagnosing conditions that impact both ears equally, such as age-related hearing loss or ototoxicity due to medications. Further, since the audio stimulus may be given to both ears simultaneously, such a technique may allow faster mapping, making this technique efficient for clinical settings where rapid screening is required.

[0085] In various embodiments, the EEG-based neural response data may be collected during both- ear stimulation. The collected data may be further analyzed for waveform characteristics such as peak amplitude, response latency, and frequency response. These metrics may be bilaterally compared to identify any discrepancies that could suggest asymmetry in auditory processing.

[0086] In another embodiment of the present disclosure, the system 100 may render single-side audio mapping. In the single-side audio mapping, the audio stimulus may be provided to one of the ears at a time, which may allow isolated testing of each ear’s auditory pathway. The same process may be repeated for another ear separately, which may create a detailed, ear-specific profile of neural responsiveness. By' isolating the stimulation to one ear, the present disclosure may effectively highlight ear-specific deficits that may not be detected with bilateral stimulation.

[0087] In an implementation, the auditory' stimulus may be provided in an alternating manner, typically with a short inter-trial interval (ITT) to allow' the neural response to return to baseline before presenting a next tone (auditory stimulus). Thus, the single-side audio mapping may be ideal for detecting unilateral hearing loss or conditions that may affect one ear disproportionately. By analyzing each ear separately, this technique may enable the identification of specific deficits, such as reduced amplitude response in one ear or prolonged latency in neural processing. Single-side mapping may further offer a detailed assessment of each auditory pathway, from the ear to the auditory cortex. This level of detail may be particularly valuable in cases where subtle asymmetries may indicate underlying neurological conditions. Additionally, the separate examination of EEG biosignals from the left and right ears may allow the study of cross-hemispheric auditory processing. This may involve evaluating how each hemisphere, i.e., the left hemisphere for the right ear and the right hemisphere for the left ear, may independently process auditory signals. This may be of particular relevance in neurological cases involving lesions or dysfunction in specific auditory' pathways.

[0088] In an implementation, each audio stimulus may be precisely directed to one of the ears at a time, which may’ often be achieved through a specialized insert in the earphones (the apparatus 102.) or circumaural headphones (the apparatus 102) that may provide effective sound isolation.

[0089] In another implementation, the EEG-based neural response data for each ear may be captured independently, with electrode placement that is designed to monitor responses in the contralateral auditory cortex. For instance, during the right-ear stimulation, the EEG response from the left temporal lobe may be closely monitored, as auditory processing is primarily contralateral. Thus, the hearing ability may be analyzed by comparing the EEG responses from each ear to detect significant discrepancies. This approach may leverage high temporal resolution to distinguish individual auditory processing events, particularly focusing on early auditory responses as well as later cortical responses that may reflect higher-order auditory’ processing.

[0090] In another implementation, sequential mapping with alternating sides may be used, where auditory’ stimuli may be presented alternately to each ear. Such an approach may enable a direct bilateral comparison of auditory processing. Alternating sides may be adapted to minimize adaptation effects and neural habituation by revealing subtle asymmetries that may be obscured in single-side or simultaneous mapping. The stimuli may be presented with a short inter-trial interval (ITI) to allow the neural response to reset, creating balanced exposure and capturing real-time ear-specific differences. Standardized audio tones or clicks, ranging from 250 Hz to 8,000 Hz, may be used to provide a comprehensive auditory spectrum profile.

[0091] In an implementation, a bilateral comparison of auditory pathways may be used for identifying asymmetries critical for diagnosing unilateral hearing loss or neural irregularities in auditory’ pathways. The alternating stimuli may’ provide comparative insights into each ear’s response. Since precise control of the interstimulus interval (ITI) is crucial for minimizing the effects of variations in response reliability, alternating stimuli ensures that each ear receives the stimulus under consistent conditions, facilitating a reliable comparison of neural responses.

[0092] In another implementation, the present disclosure may incorporate adaptive testing for individual variability’ and may accommodate individual variability by minimizing habituation, which is useful in detecting conditions with asymmetric auditory processing.

[0093] In another implementation, the present disclosure may’ incorporate an enhanced cross-ear analysis that may assess each ear independently, facilitating nuanced detection of auditory pathway discrepancies relevant to specific conditions, like processing disorders.

[0094] In an embodiment, the wearable audio output apparatus 102 may include a dedicated microphone 218 coupled to the body, the microphone 218 being adapted to capture environmental sound and to detect otoacoustic emissions (OAEs). The microphone 218, called Otoacoustic emission microphones, may be tuned for ultra-low internal noise to record faint cochlear emissions, including spontaneous OAEs (SOAEs) and evoked OAEs (EOAEs). To elicit EOAEs, the apparatus 102 may emit calibrated stimuli such as clicks or paired pure tones and record resulting transient-evoked OAEs (TEOAEs) and distortion-product OAEs (DPOAEs), thereby providing a non-invasive measure of cochlear integrity useful for early detection and differentiation of conductive versus sensorineural hearing loss. This data can be used as baseline reference to improve the EEG based Auditory Evoked Potential based hearing profile mapping of the user, as well as work in tandem with it, there by improving the overall accuracy.

[0095] In another embodiment oriented to cerebrovascular moni toring, an ultra low-frequency tuned microphone may capture tissue-bome acoustic signatures transmitted through the ear canal, including pulse-related acoustic signals, heart rate variability components, and other low-amplitude vascular sounds. The apparatus 102 periodically or continuously analyses these acoustic waveforms for asymmetries, diminished pulsatile energy, or abnormal rhythmic patterns that may indicate restricted blood flow or cerebrovascular anomalies, and may combine such detections with alerting logic for early warning of ischemic events.

[0096] In another embodiment, the apparatus 102. may further include a vascular-coupled physiological sensing module that employ non-acoustic transducers such as a micro-accelerometer embedded in the earbud housing, bioimpedance sensing electrodes on the earbud surface, or ear-EEG electrode structures, the sensors being configured to detect tissue-conducted micro-vibrations, dynamic impedance changes, or electrophysiological correlates of local perfusion and pulsatile blood volume.

[0097] In one implementation, a micro-accelerometer senses minute mechanical accelerations transmitted through cartilage and bone to extract pulse timing, waveform morphology, and vascular stiffness indicators; in another implementation, a low-magnitude excitation for bioimpedance sensing measures dynamic impedance variations correlated with local blood perfusion and vascular compliance. Owing to the earbud’s dynamically adjusting electrode and housing design, the sensor assembly can achieve a more stable and conformal mechanical interface with the walls of the ear canal. This adaptive sealing improves coupling efficiency with the close contact and reduces motion artefacts compared to sensors mounted on non-adaptive or externally fixed structures, thereby enhancing the fidelity and reliability of the captured hemodynamic and electrophysiological signals.

[0098] The apparatus 102 may integrate mechanical or impedance-derived hemodynamic features with concurrently acquired EEG, ECG or EMG to evaluate neurovascular coupling, changes in evoked response latency, autonomic rhythm asymmetry, or transient disruptions in cerebral electrophysiology. Deviations from an established per-user baseline across these multimodal signals may be analysed to identify early signatures of cerebrovascular compromise, elevated stroke risk, or abnormal autonomic patterns, potentially alerting the user, an ambulance, or a medical professional in real-time. This passive, non-invasive monitoring approach supports long-term cerebrovascular health tracking, offering the potential for early intervention and continuous stroke risk assessment without the need for active user involvement.

[0099] Now, the system 100 of the present disclosure may be used to diagnose various hearing- related defects associated with the user, which is discussed in the forthcoming paragraphs.

[0100] The system 100 may be capable of diagnosing presbycusis (age-related hearing ioss). In an embodiment, the system 100 may track age-related hearing loss by monitoring Auditory brainstem response (ABR) to understand the degradation over time. The system 100 may analyze changes in latency and waveform stability in response to different frequencies, thereby enabling early detection and ongoing assessment of the presbycusis, guiding compensatory interventions to maintain sound clarity.

[0101] The system 100 may be capable of diagnosing tympanic membrane abnormalities. In an embodiment, the system 100 may provide targeted diagnostics for perforations or structural defects in the tympanic membrane by analyzing auditory response changes and identifying disruptions in sound conduction to the middle ear. In an exemplary embodiment, the microphone 218 may be placed within the ear canal to assist in diagnosis. This may aid in detecting and evaluating the severity of tympanic membrane issues, supporting precise treatment decisions.

[0102] In various embodiments, the system 100 may detect cerumen (earwax) impaction. For cases of earwax buildup causing temporary conductive hearing loss, the system 100 may offer a non- invasive way to detect blocked sound transmission. In an advantageous effect, the EEG-based detection may inform timely diagnosis and guide interventions, such as earwax removal, to restore hearing.

[0103] In various embodiments, the system 100 may detect Benign Paroxysmal Positional Vertigo (BPPV). In an embodiment, the system 100 may assess auditory-vestibular interaction in cases of the BPPV, where vertigo is triggered by head movements due to displaced inner ear particles. In anembodiment, the EEG-based neural response data in response to specific stimuli may help track auditory’ shifts and recovery patterns, aiding diagnosis and supporting comprehensive management.

[0104] In an embodiment, the system 100 may help in cochlear implant optimization. In an embodiment, for cochlear implant users, the system 100 may aid in tuning and personali zing auditory settings by analyzing auditory pathway response delays and waveform clarity. The system 100 provides EEG-based insights into response effectiveness, enabling fine-tuning that enhances hearing outcomes and auditory experiences.

[0105] Now, the advantages of the present disclosure are discussed in the forthcoming paragraphs.

[0106] The present disclosure enables integration of the biosignal electrodes 104, such as the EEG sensors, directly into consumer-grade earphones (the wearable audio output apparatus 102), thereby creating an adaptive audio interface capable of autonomously compensating forbearing deficiencies. Unlike traditional hearing aids that rely on external amplification or preset filters, the earphones (the wearable audio output apparatus 102) continuously monitor the EEG-based neural response data through the EEG-based electrodes (biosignal electrodes 104) and dynamically adjust playback characteristics to maintain perceptual clarity. This combination enables real-time neuroadaptive gain control and frequency correction without the need for medical grade hearing aids, offering a non¬ stigmatizing and seamless hearing enhancement solution within a familiar consumer device form factor. Further, the present disclosure enables adjustment of the audio output based on the hearing capabilities of the user, thereby aligning the audio output with the users real-time neural response, i.e., with a personalised frequency response profile, thereby enabling delivery of the user-specific audio output to each user.|00107] Further, the present disclosure utilizes the EEG in hearing assessments may offer significant ad vantages over traditional methods, particularly in the detection of sensorineural hearing loss and early-stage auditory degeneration. The present disclosure enables capturing subtle changes m neural activity that reflect early signs of auditory nerve deterioration or cortical processing deficits, which might go undetected with conventional methods. Thus, to accurately test hearing ability, the EEG biosignals may be used to capture real-time neural responses to the auditory’ stimulus, which may allow for continuous and dynamic monitoring of auditory-’ processing.

[0108] In one embodiment, the present disclosure employs a machine learning framework to enhance individualized auditory processing and frequency-specific hearing assessment by leveraging asynchronous federated learning Al models. Each user's device may run a local machine learning model, possibly using edge Al technology, to continuously analyze personal EEG responses andauditory data, estimating frequency-specific hearing deficiencies and the response patterns. These local models are periodically recalibrated based on new invasive calibration data and evolving EEG patterns, ensuring personalized accuracy in assessing hearing sensitivity. To further refine overall accuracy, a cloud-based global model aggregates anonymized EEG and auditory response data from multiple users, allowing the system 100 to identify broader trends that may not be evident in individual datasets.

[0109] The present disclosure incorporates advanced ML techniques, such as large language models (LLMs) and convolutional neural networks (CNNs), to interpret the EEG and auditory response data collected from users. Local devices run CNN-based models to detect patterns in EEG responses to specific frequencies, while LLMs provide context-aware interpretation of these responses. The local models are updated periodically with new calibration data, refining each user’s frequency response profile and enhancing personalized audio adjustments. This dual-model framework combines real- time EEG analysis with deep learning, optimizing frequency -specific volume adjustments to match each user’s unique hearing profile.

[0110] In an alternative embodiment, the system 100 integrates CNNs and deep learning models to analyze EEG signals and auditory response data for identifying frequency-specific hearing sensitivity. These models are designed to continuously monitor real-time EEG responses to audio stimuli, updating user profiles based on calibration data. Tins process allows the system 100 to maintain highly accurate, personalized frequency-specific hearing assessments, adapting dynamically to any shifts in the user's auditory perception over time.

[0111] In another embodiment, the present disclosure includes a federated learning approach for decentralized data processing, ensuring user data remains private while benefiting from large-scale analysis across users. Federated learning enables the system 100 to aggregate anonymized hearing response data, identifying population -wide patterns while maintaining privacy. Through the Edge Al architecture, each device processes data locally, minimizing latency and avoiding the need for continuous cloud communication. This structure improves processing efficiency and allows the system 100 to deliver real-time feedback, enhancing both response accuracy and speed in hearing diagnostics.

[0112] In a further embodiment, the system 100 uses transformer-based architectures, such as large language models (LLMs), to interpret complex auditory- and the EEG neural response data in a context-aware manner. By capturing nuanced patterns in the EEG neural response data to various frequencies, the LLMs provide a more comprehensive view of the user’s hearing profile. Thesemodels, updated with both personalized and aggregated data, remain adaptive and robust, enabling effective frequency-specific volume adjustments and guiding customized hearing compensation strategies over time. This advanced processing framework facilitates precise and dynamic hearing management, ensuring the listener’s (the user) auditory experience closely aligns with their individual frequency response profile.

[0113] Further, the present disclosure, by analyzing the EEG responses in response to the auditory stimulus, the system 100 may correlate neural activity with auditory perception, enabling precise mapping of auditor}' nerve function. The system’s 100 hardware and software components are integrated to facilitate the correlation of audio to the EEG responses, making it user-friendly and suitable for clinical and at-home applications.

[0114] While specific language has been used to describe the present subject matter, any limitations arising on account thereto, are not intended. As would be apparent to a person in the art, various working modifications may be made to the method in order to implement the inventive concept as taught herein. The drawings and the foregoing description give examples of embodiments. Those skilled in the art will appreciate that one or more of the described elements may well be combined into a single functional element. Alternatively, certain elements may be split into multiple functional elements. Elements from one embodiment may be added to another embodiment.

Claims

WE CLAIM1. A wearable audio output apparatus (102) for a user, comprising:a body adapted to be worn proximal to at least one ear of the user, the body comprising one or more biosignal electrodes (104) coupled thereto; anda processor (302) in communication with the one or more biosignal electrodes (104), the processor (302) configured to:receive audio input from a user device (106) associated with the user; identify auditory' stimulus from the audio input, wherein the auditory' stimulus being indicative of one or more biomarkers that trigger a neural response from the brain of the user;obtain, from the one or more biosignal electrodes (104), electroencephalogram (EEG)-based neural response data corresponding to the auditory stimulus; identify a frequency range associated with altered response patterns based on the (EEG)-based neural response data;modulate amplitude corresponding to the frequency range; and deliver, via a speaker unit (214), a user-specific audio output to the user based on the modulation.

2. The wearable audio output apparatus (102) as claimed in claim 1 comprising a microphone (218) coupled to the body, wherein the microphone (218) is adapted to capture environmental sound.

3. The wearable audio output apparatus (102) as claimed in claim 1, wherein the processor (302) is one of wired or wirelessly connected with the user device (106).

4. The wearable audio output apparatus (102) as claimed in claim 1, wherein to identify the frequency range associated with the altered response patterns based on the EEG-based neural response data, the processor (302) is configured to:analyze response patterns of the EEG-based neural response data;detect, using an artificial intelligence (Al) model, the altered response patterns based on the analyzing, andidentify, using the Al model, the frequency range associated with the altered response patterns.

5. The wearable audio output apparatus (102) as claimed in claim 1, wherein to modulate the amplitude corresponding to the frequency range, the processor (302) is configured to:generate, using adaptive equalizer algorithm, a personalized frequency response profile corresponding to the user; andmodulate the amplitude corresponding to the identified frequency range within the personalized frequency response profile.

6. The wearable audio output apparatus (102) as claimed in claim 5, wherein the processor (302) is configured to:update the personalized frequency response profile in real time based on variations in the EEG-based neural response data corresponding to the auditory' stimulus;dynamically refining audio output characteristics including at least one of gain, tone, and frequency response, such that a synchronization is maintained with the user’s neural response; andselectively one of increasing or decreasing amplitudes of one or more frequency components to facilitate adaptive auditory- conditioning.

7. The wearable audio output apparatus (102) as claimed in claim 6, wherein the processor (302) is configured to identify gradual shifts in auditor}' perception arising from one or more factors using a machine learning (ML) model, wherein the one or more factors comprises at least one of aging, environmental changes, or neural adaptation.

8. The wearable audio output apparatus (102) as claimed in claim 1, wherein the processor (302) operates within a closed-loop adaptive framework, the processor (302) is configured to: obtain neurofeedback data corresponding to the one or more users from the user device (106) associated with each of the one or more users;refine audio adjustment parameters based on the neurofeedback data; anddetect, using the refined audio adjustment parameters, altered auditory response patterns from the EEG-based neural response data.

9. The wearable audio output apparatus ( 102) as claimed in claim 1, wherein the processor (302) is further configured to:resolve audio ambiguity by leveraging crowdsourced neurofeedback data obtained through the federated learning architecture, wherein to resolve the audio ambiguity, the processor (302) is configured to:identify targeted audio segments based on crowdsourced neurofeedback data, the targeted audio segments are indicative of at least one of weak or inconsistent EEG engagement patterns across the crowdsourced neurofeedback data, and reconstruct, using neural patching models, the targeted audio segments, thereby generating neuro-tuned audio output.

10. A method (400) for delivering user-specific audio output to a user, the method (400) comprising:receiving audio input from a user device (106) associated with the user; identifying auditory stimulus from the audio input, wherein the auditory stimulus being indicative of one or more biomarkers that trigger a neural response from the brain of the user;obtaining, from the one or more biosignal electrodes (104), electroencephalogram (EEG)-based neural response data corresponding to the auditory stimulus;identifying a frequency range associated with alerted response patterns based on the (EEG)-based neural response data;modulating amplitude corresponding to the frequency range; anddelivering, via a speaker unit (214) of a wearable audio output apparatus (102), the user¬ specific audio output to the user based on the modulation.

11. The method (400) as claimed in claim 10, wherein identifying the frequency range associated with the alerted response patterns based on the EEG-based neural response data comprises:analyzing response patterns of the EEG-based neural response data;detecting, using an artificial intelligence (Al) model, the alerted response patterns based on the analyzing, andidentifying, using the Al model, the frequency range associated with the alerted response patterns.

12. The method (400) as claimed in claim 10, wherein modulating the amplitude corresponding to the frequency range comprises:generating, using adaptive equalizer algorithm, a personalized frequency response profile corresponding to the user; andmodulating the amplitude corresponding to the identified frequency range within the personalized frequency response profile.

13. A system (100) for delivering user-specific audio output to a user, the system (100) comprising:a wearable audio output apparatus (102) adapted to be worn by the user, proximal to at least one ear of the user;one or more biosignal electrodes (104) coupled to the wearable audio output apparatus (102); anda processor (302) in communication with the one or more biosignal electrodes (104) and the wearable audio output apparatus (102), the processor (302) configured to:receive audio input from a user apparatus (102) associated with the user; identify auditory stimulus from the audio input, wherein the auditory' stimulus being indicative of one or more biomarkers that trigger a neural response from the brain of the user;obtain, from the one or more biosignal electrodes (104), electroencephalogram (EEG)-based neural response data corresponding to the auditory stimulus;identify’ a frequency range associated with alerted response patterns based on the (EEG)-based neural response data;modulate amplitude corresponding to the frequency range; and deliver, via a speaker unit (214), the user-specific audio output to the user based on the modulation.

Images