System and method for automated detection of headphone placement for an audiometric test

Abstract

There is provided a system (100, 200, 300, 400) for automated detection of a placement of a headphone (108, 216, 302, 412) on an ear of a user (110, 218, 414) for an audiometric test. The system comprises a camera (106, 202, 306), the headphone, and an image processing unit (102, 220, 406, 500, 600, 703). The camera obtains imaging data of the headphone coupled to the ear of the user. The headphone comprises a marker (114A, 114B). The headphone is configured to be used for an audiometric test. The image processing unit processes the imaging data to automatically detect the marker. Based on the detected marker, the image processing unit determines if a first placement of the headphone on the ear of the user corresponds to a predetermined placement of the headphone to perform the audiometric test. Based on the determination, the image processing unit is configured to provide a feedback signal to the user, reconfigure an audio channel of the headphone automatically, or monitor the obtained imaging data during the audiometric test.

Description

[0001] SYSTEM AND METHOD FOR. AUTOMATED DETECTION OF HEADPHONE PLACEMENT FOR AN AUDIOMETRIC TEST

[0002] TECHNICAL FIELD

[0003] The present disclosure relates to the field of audiometric testing. More particularly, the present disclosure relates to imaging technologies for detecting headphone placement. The disclosure further relates to technologies associated with the use of headphones in clinical and medical examination environments for hearing assessment.

[0004] BACKGROUND

[0005] Audiometric testing is a process for evaluating an individual's hearing capability by determining audible thresholds across a range of sound frequencies. During audiometric testing, acoustic test tones at varying intensity levels are presented to the individual, and the individual's responses are recorded to assess hearing sensitivity. The results of such testing are typically represented in an audiogram, which provides a graphical representation of audible thresholds as a function of frequency. Audiograms are used for diagnosing hearing loss, determining its type and severity, and configuring hearing aids.

[0006] Conventional audiometric testing methods are time-consuming and require specialised equipment operated by trained professionals. Individuals are required to schedule appointments at clinical facilities, wait for testing, and incur relatively high costs. Further, hearing aids configured on the basis of conventional audiometric testing may not fully address an individual's specific auditory needs, which can result in suboptimal hearing performance.

[0007] Many existing audiometric testing systems employ headphones to deliver test tones independently to a left ear and a right ear of a user. However,such systems do not ensure that the headphones are correctly positioned on the user's ears during testing. For example, the headphones may be worn in a reversed orientation, such that test signals intended for the left ear are delivered to the right ear and vice versa. Incorrect positioning of the headphones may result in erroneous audiograms, inaccurate diagnosis of hearing conditions, and improper configuration of hearing aids. In addition, existing audiometric testing methods do not monitor physiological or contextual parameters that could provide information regarding the reliability of a user's responses during the testing process.

[0008] Certain known headphone systems include a speaker, a physiological sensor for measuring physiological data, and a processor configured to process the measured data to generate physiological information and a fitting parameter. Such headphone systems may evaluate the fitting parameter using an application to indicate whether the headphone is properly positioned on a user's ears. However, these systems are related only to headphone fitting and do not address the detection of incorrect or reversed placement of left and right earpieces during audiometric testing.

[0009] Other known systems disclose one or more earpieces that are configured to perform audiometric testing. The earpiece includes an earpiece housing, an intelligent control system, at least one transducer, and at least one speaker. During operation, test sounds are reproduced through the transducer, and user feedback is received to conduct audiometric testing. However, such known systems do not detect improper placement or reversal of the earpiece on a user's ears during the testing process.

[0010] Therefore, in light of the foregoing discussion, there is a need to overcome the aforementioned limitations to enhance the consistency, reliability, and efficiency of the audiometric test in clinical practice.SUMMARY

[0011] The aim of the present disclosure is to provide a method, a system, and a computer-readable storage medium for automated detection of a placement of a headphone on an ear of a user for an audiometric test. The aim of the disclosure is achieved by a method, a system, and a computer-readable storage medium for automated detection of a placement of a headphone on an ear of a user for an audiometric test, as defined in the appended independent claims to which reference is made. Advantageous features are set out in the appended dependent claims.

[0012] The embodiments of the present disclosure substantially enable an uninterrupted and more accurate audiometric test by automatically detecting and correcting erroneous audio channel assignment caused by incorrect headphone placement using a computer vision technique.

[0013] Additional aspects, advantages, features, and objects of the present disclosure are made apparent from the drawings and the detailed description of the illustrative embodiments construed in conjunction with the appended claims that follow.

[0014] Throughout the description and claims of this specification, the words "comprise", "include", "have", and "contain" and variations of these words, for example "comprising" and "comprises", mean "including but not limited to", and do not exclude other components, items, integers or steps not explicitly disclosed also to be present. Moreover, the singular encompasses the plural unless the context otherwise requires. In particular, where the indefinite article is used, the specification is to be understood as contemplating plurality as well as singularity, unless the context requires otherwise.BRIEF DESCRIPTION OF THE DRAWINGS

[0015] FIG. 1 is a schematic illustration of a system for automated detection of a placement of a headphone on an ear of a user for use in an audiometric test in accordance with an embodiment of the present disclosure;

[0016] FIG. 2 is an illustration of a system for automated detection of a placement of a headphone on an ear of a user for use in an audiometric test in accordance with an embodiment of the present disclosure;

[0017] FIG. 3 is an illustration of an example dock that supports a relative positioning of components during an audiometric test, representing a system in accordance with an embodiment of the present disclosure;

[0018] FIG. 4 is a schematic illustration of a system for automated detection of a placement of a headphone on an ear of a user for use in an audiometric test in accordance with an embodiment of the present disclosure;

[0019] FIG. 5 is a schematic illustration of an image processing unit of a system in accordance with an embodiment of the present disclosure;

[0020] FIG. 6 is a schematic illustration of an image processing unit of a system in accordance with an embodiment of the present disclosure;

[0021] FIG. 7 is an illustration of an architecture of a system showing an interaction between software components and hardware components for an audiometric test in accordance with an embodiment of the present disclosure;

[0022] FIG. 8 illustrates a flowchart of a method for automatically detecting and correcting headphone placement during an audiometric test in accordance with an embodiment of the present disclosure;FIG. 9 illustrates a flow diagram of a method for automated detection of a placement of a headphone and indicators of a user for an audiometric test, in accordance with an embodiment of the present disclosure;

[0023] FIG. 10 is an illustration of a flowchart illustrating steps of a method for (namely, a method of) automated detection of a placement of a headphone on an ear of a user for an audiometric test in accordance with an embodiment of the present disclosure; and

[0024] FIG. 11 is an illustration of a block diagram of a computing architecture / system according to an embodiment of the present disclosure.

[0025] In the accompanying drawings, an underlined number is employed to represent an item over which the underlined number is positioned or an item to which the underlined number is adjacent. A non-underlined number relates to an item identified by a line linking the non-underlined number to the item. When a number is non-underlined and accompanied by an associated arrow, the non-underlined number is used to identify a general item at which the arrow is pointing.

[0026] DETAILED DESCRIPTION OF EMBODIMENTS

[0027] The following detailed description illustrates embodiments of the present disclosure and ways in which they can be implemented. Although some modes of carrying out the present disclosure have been disclosed, those skilled in the art would recognize that other embodiments for carrying out or practicing the present disclosure are also possible.

[0028] In a first aspect, the present disclosure provides a computer-implemented method for automated detection of a placement of a headphone on an ear of a user for an audiometric test, the method comprising:obtaining imaging data of the headphone coupled to the ear of the user, wherein the headphone comprises a marker;

[0029] processing, using an image processing, the imaging data to automatically detect the marker;

[0030] determining, based on the detected marker, if a first placement of the headphone on the ear of the user corresponds to a predetermined placement of the headphone to perform the audiometric test; and

[0031] based on the determination, performing at least one of the following:

[0032] - providing a feedback signal to the user,

[0033] - reconfiguring automatically an audio channel of the headphone, or

[0034] - monitoring the obtained imaging data during the audiometric test.

[0035] The method enables automatic detection and correction of incorrect or reversed headphone placement without manual intervention, thereby maintaining uninterrupted test conditions, and reducing errors in audiometric measurement. Automatic reconfiguration of the audio channel ensures that the audio channel is delivered to the intended ear, thereby improving the accuracy of the audiometric data while maintaining continuity. The method further enables monitoring of user-related parameters observable from the imaging data, such as head movement, posture, or visible indicators of discomfort or fatigue, thereby enabling an improved assessment of test conditions during an audiometric test.

[0036] The method provides a non-invasive and automated approach for detecting headphone placement using computer vision and imageprocessing techniques, thereby reducing manual intervention by an operator and reducing human error during the audiometric test.

[0037] The imaging data may comprise one or more still images and / or video frames captured by an image capturing unit (e.g., a camera) that is configured to obtain a view of the user while the headphone is worn. The image capturing unit may be positioned at a predetermined distance and orientation relative to the user's head so as to capture the headphone and ear region with high resolution, thus having controlled imaging geometry via dock to improve capture repeatability and to reduce variance, at the same time improving detection robustness. In an embodiment, the image capturing unit is integrated into a laptop, desktop computer, smartphone, tablet, or other user device.

[0038] The imaging data may be processed using an image processing technique to automatically detect the marker provided on the headphone. The marker may be arranged on a pre-defined portion of an earcup or earpiece and is configured to distinguish between left and right orientations of the headphone. The detected marker is analyzed to identify the placement of the headphone relative to the user's ears, including whether a left earpiece is positioned on a left ear or a right ear, and vice versa.

[0039] The image processing may comprise computer vision-based analysis employing one or more of an object detection technique, a feature-based detection technique, a pattern recognition, a colour thresholding and segmentation, or a template matching. In an embodiment, the imaging data is first subjected to preprocessing operations, such as noise reduction, contrast or brightness adjustment, image cropping or resizing, edge enhancement, or spatial filtering. These preprocessing operations improve the detection of the marker by reducing noise and correcting variations in illumination or background conditions.The feature-based detection techniques may be employed to identify and track one or more visual features associated with the marker or surrounding structures, such as edges, corners, contours, or landmarks. The extracted visual feature may be encoded as numerical descriptors representing geometric or structural characteristics. By analysing the visual feature across consecutive image frames, the placement and orientation of the headphone can be determined, even in the presence of user movement, partial occlusion of the marker, or changing lighting conditions.

[0040] Based on the detected marker, the method determines whether a first / current placement of the headphone corresponds to a predetermined placement suitable for performing the audiometric test. The first / current placement refers to a physical position and orientation of the headphone relative to the user's ear at the time the imaging data is captured. The predetermined placement corresponds to a reference configuration in which the headphone is correctly positioned, for example, with a left earpiece on the left ear and a right earpiece on the right ear, within the defined position that is required for an accurate audiometric test. The configuration of the predetermined placement may be stored in a memory accessible to the image processing unit.

[0041] Based on the determination, at least one action comprising (a) providing a feedback signal to the user, (b) reconfiguring automatically an audio channel of the headphone, or (c) monitoring the obtained imaging data during the audiometric test, is performed. In some embodiments, the feedback signal is provided to the user or an operator. The feedback signal may comprise a visual indication displayed on a screen, an audio prompt, or both, which provides information regarding the adjustment of the headphone placement. In an embodiment, the audio channel of the headphone is automatically reconfigured or swapped to compensate for reversed / incorrect placement, thereby enabling the audiometric test toproceed without requiring physical realignment of the headphone. The imaging data may be monitored continuously during performance of the audiometric test to confirm that the headphone remains in the predetermined placement throughout the test duration.

[0042] In an embodiment, the determination of the headphone placement is represented by a Boolean indicator or similar logical flag. For example, a first value may indicate that the headphone placement corresponds to the predetermined placement required for the audiometric test, while a second value may indicate incorrect or reversed placement.

[0043] Optionally, the automatic reconfiguration of the audio channel comprises at least one of (i) reconfiguring a first audio channel intended for a left earpiece to a right earpiece of the headphone positioned on a left ear, or (ii) reconfiguring a second audio channel intended for the right earpiece to the left earpiece of the headphone positioned on a right ear.

[0044] The audio channel for the headphone refers to a signal path configured to provide an audio signal to the earpiece of the headphone. Each audio channel is operable to transmit sound corresponding to the left or right audio channel. The reconfiguration of the audio channel of the headphone compensates for the reversed / incorrect placement of the headphone, thereby enabling the audiometric test to proceed without requiring physical realignment of the headphone.

[0045] Optionally, the method further comprises

[0046] obtaining facial image data of the user that is captured using an image capturing unit for the audiometric test;

[0047] detecting one or more indicators of the user by processing the facial image data using a facial analysis;identifying a first condition when the one or more indicators that are detected persist during the audiometric test;

[0048] automatically pausing the audiometric test in response to the identified first condition and generating a notification to at least one of an operator or the user;

[0049] identifying a second condition when the one or more indicators that are detected persist during the audiometric test; and

[0050] resuming the audiometric test upon determining that the user has returned to the second condition.

[0051] This enables identification and handling of user states that may affect response reliability, thereby improving consistency and validity of audiometric test results.

[0052] In an embodiment, the method implements a feature-based detection technique to detect head motion of the user by identifying and tracking one or more key points across successive image frames of the facial image data. Such key points may include, for example, one or more facial landmarks associated with the eyes, nose, mouth, or jaw, and are tracked over time to determine head posture and head movement.

[0053] The facial image data may be obtained using the image capturing unit positioned to capture the user's face while the headphone is worn by the user. The image capturing unit may be positioned at a predetermined angle and at a predetermined distance relative to the user's head in order to obtain facial image data. The facial image data may be analysed during the audiometric test to detect one or more indicators, including facial cues, eye-related cues, head posture cues, and head movement cues. The detected indicators may be analysed to determine whether the first condition is present. In an embodiment, the first condition correspondsto a state of the user indicative of discomfort, fatigue, stress, reduced attentiveness, or difficulty in perceiving certain sound frequencies or sound intensity levels. Such a state may reflect a physical or cognitive condition that could adversely affect the audiometric test results. The facial cues may include frowning, brow furrowing, squinting, or jaw tension. When the first condition is identified based on the facial cues, the audiometric test is automatically paused, and a prompt is generated for the user.

[0054] In another example, the eye-related cues may include changes in pupil dilation indicative of stress or discomfort, an increased blink rate, prolonged eye closure indicative of fatigue, or reduced gaze stability. Upon detecting the first condition based on the eye-related cues, the audiometric test is automatically paused, and a prompt is generated, for example, instructing the user to take a short break.

[0055] In an embodiment, changes in the head posture cues or head movement cues are detected during the audiometric test. Such changes may include frequent head turning, repeated repositioning, or slouching, which may indicate discomfort, confusion, or fatigue. Upon identifying the first condition based on the head posture or head movement cues, the audiometric test is paused, and the user is prompted to reposition before continuation of the test.

[0056] In another embodiment, upon identification of the first condition, sound intensity for the left and right earpieces of the headphone is automatically reduced, for example, by reverting to a lower-level test tone, in order to improve user comfort and enable more reliable test conditions. The second condition may correspond to a relaxed or stable state of the user, which is determined from the detected indicators, upon which the audiometric test is automatically resumed.Optionally, the marker comprises a pattern, a colour, an alphanumeric code, or a symbol, indicating whether the earpiece is the left earpiece or the right earpiece. Optionally, the marker is an Infrared (IR) reflective. This enables reliable differentiation between left and right earpieces in the imaging data under varying environmental and illumination conditions, thereby supporting accurate determination of headphone placement during the audiometric test.

[0057] For example, the left earpiece may be provided with a first marker, such as a first symbol or pattern, and the right earpiece may be provided with a second marker that is visually distinct from the first marker. This arrangement enables differentiation between the left and right earpieces of the headphone during image-based detection.

[0058] The method may apply a colour-based detection technique, including colour thresholding and segmentation, when the marker has visually distinguishable colours associated with respective earpieces. For example, a first colour may be associated with the left earpiece and a second colour may be associated with the right earpiece. For example, the first colour is red, and the second colour is blue, or vice versa. The thresholding and segmentation technique enables efficient detection under controlled illumination conditions.

[0059] The method may employ a template matching technique, such as normalized cross-correlation, when the markers comprise high-contrast visual patterns or symbols. Under consistent lighting conditions and limited occlusion, the template matching technique may provide a computationally efficient mechanism for detecting and distinguishing the markers associated with the respective earpieces.

[0060] In an embodiment, the method employs a feature-based detection technique, such as Scale-Invariant Feature Transform (SIFT), Speeded-Up Robust Features (SURF), or Oriented FAST and Rotated BRIEF (ORB),when the earpieces include different visual features, shapes, or textured patterns.

[0061] Optionally, the method comprises positioning the image capturing unit at a distance of between 0.5 meter (m) and 3 meter (m) in front of the user and at an angle between 30° and 75° relative to a frontal plane of the user so that the earpiece is visible in the imaging data. This positioning on the setup turns "computer vision on a generic laptop" into a controlled measurement setup with fixed geometry, repeatable FOV, reduced variance, and positioning ensures consistent visibility of the earpiece in the imaging data, thereby improving robustness and reliability of placement detection during the audiometric test. Moreover, in practical implementations, the image capturing unit may be configured to provide different horizontal fields of view depending on a desired framing of the user, including a close-up framing in which a head-and-shoulders region substantially fills an image frame with a horizontal field of view of approximately 50 to 60 degrees, a mid-range framing capturing a head-to-chest region with a horizontal field of view of approximately 60 to 75 degrees to accommodate user movement or misalignment. This configuration further reduces the volume of imaging data to be processed for placement detection, thereby lowering computational load and latency while maintaining sufficient visual coverage for reliable operation. Optionally a wider environmental framing capturing may be used for additional surroundings or multiple persons with a horizontal field of view of approximately 75 to 90 degrees.

[0062] Optionally, the image capturing unit captures the imaging data at a frame rate of 10 to 30 frames per second. The frame rate is used to detect changes in the placement of the headphone without a delay. The image capturing unit / camera may capture the imaging data at a resolution of at least 1280 x 720 pixels, which enables accurate detection of the marker on the headphone. This frame rate enables timely detection of changesin headphone placement and user movement, thereby supporting continuous and responsive monitoring during the audiometric test. Capturing imaging data at 10-30 frames per second provides sufficient temporal resolution for reliable marker-based pose estimation and placement verification while reducing compute load, memory bandwidth, and power consumption compared to higher frame rate capture. This frame rate range also reduces detection latency and improves correction responsiveness compared to lower frame rate capture, improving overall reliability and throughput.

[0063] Optionally, the image processing is performed using a machine learning model that is trained with at least one of:

[0064] (i) a dataset of images with correctly and incorrectly placed headphones, markers, subtle facial landmarks, head pose, skin tone, and headphone material;

[0065] (ii) an augmented dataset that comprises images in low-light conditions, stand, user's chair, and additional data from one or more sensors of the headphone; or

[0066] (iii) a dataset that comprises different headphone textures or images of users with different demographic characteristics.

[0067] The image processing for detecting the placement of the headphone may be performed using the machine learning model that is trained to classify correct and incorrect placement of the headphone on the user's ear. The machine learning model may be implemented, for example, as a convolutional neural network (CNN), a transformer-based vision model, or a hybrid architecture combining learned and rule-based features.

[0068] The machine learning model may be trained using one or more datasets comprising images of users wearing headphones in both correctly and incorrectly placed configurations. Such datasets may include visualrepresentations of markers on the headphone, facial landmarks of the user, head pose information, variations in skin tone, and differences in headphone materials or finishes. These features enable the machine learning model to learn correlations between visual cues and headphone placement orientation.

[0069] The training dataset may be augmented with synthetically generated or captured images representing low-light conditions, different backgrounds such as a testing booth or user chair, and additional data obtained from one or more sensors integrated into the headphone, such as inertial or proximity sensors. Further, the dataset may comprise images of headphones with different surface textures or images of users having different demographic characteristics. This improves the accuracy of detection across a user population. Further, the use of trained machinelearning models improves accuracy over rule-based approaches and reduces noise, occlusion, and lighting variability, thereby enhancing the accuracy of the audiometric test.

[0070] Optionally, the imaging data of the headphone or the facial image data of the user is obtained using at least one of: an infrared camera, a depth sensor, a hyperspectral detector, or a thermal camera for detecting the placement of the headphone under varying lighting conditions, or partial occlusion. This enables detection of headphone placement under low-light conditions, partial occlusion, or varying environmental factors, thereby improving robustness of placement determination. It also improves placement detection robustness under low / variable ambient lighting and partial occlusion by providing additional non-visible or geometric signal channels beyond RGB imaging. This reduces false detections and the need for repeated image capture or reprocessing, improving correction latency and overall test reliability compared to conventional visible-light camera-only solutions.In an embodiment, the infrared camera may be used to detect markers or reflective elements on the headphone under low-light or uneven illumination conditions. The depth sensor, selected from structured-light sensors or time-of-flight sensors, may be used to acquire three-dimensional information representing a position of the headphone with respect to the user's ear and head geometry. The hyperspectral detector may capture wavelength-specific reflectance characteristics of the marker or headphone material, and the thermal camera may detect heat patterns associated with skin contact or proximity.

[0071] These sensors may be used individually or in combination to provide information for determining headphone placement, particularly in conditions involving partial occlusion, shadows, or reflective glare.

[0072] Optionally, when the marker of the headphone is at least partially blocked from view in the imaging data of the headphone, the method comprises identifying one or more visible features in the imaging data using a feature-based image processing or a convolutional neural network; and

[0073] determining, based on the one or more identified visible features, if the first placement of the headphone on the ear of the user corresponds to the predetermined placement of the headphone to perform the audiometric test.

[0074] In an embodiment, the identified visible features include geometric contours of the earpiece, a position of the headphone band, ear shape, facial landmarks, or symmetry characteristics of the user's head.

[0075] Based on the identified visible features, the method determines whether the detected placement of the headphone corresponds to the predetermined placement suitable for the audiometric test. The determination may involve probabilistic inference or confidence scoringderived from the detected visible features. This allows accurate placement determination even when direct visual access to the marker is impaired.

[0076] Optionally, the method comprises

[0077] detecting a lighting condition within the imaging data of the headphone or the facial image data of the user;

[0078] determining whether the lighting condition is below a pre-defined threshold lighting condition; and

[0079] based on the determination, performing at least one of the following:

[0080] switching automatically from the detected lighting condition to infrared-based tracking, wherein the infrared-based tracking is performed using an infrared camera,

[0081] adjusting automatically one or more imaging parameters of the image-capturing unit, based on the pre-defined threshold lighting condition, or

[0082] generating a prompt to the operator to adjust the lighting condition.

[0083] This automatic adjustment of the lighting condition enables continuous, adaptive and reliable placement detection under dynamically changing illumination conditions, thereby maintaining robustness of the audiometric test under adverse conditions. This reduces operator intervention, avoids test interruptions, and ensures consistent data quality during the audiometric test. The one or more imaging parameters may include exposure, white balance, shutter speed, aperture, and focus. The pre-defined threshold lighting condition refers to a pre-defined value or range of ambient light intensity. Detecting insufficient lighting andautomatically switching to infrared tracking and / or adjusting capture parameters reduces placement-detection errors and re-capture cycles under low-light conditions, thereby improving tracking accuracy and correction latency compared to fixed-configuration visible-light imaging systems. Detecting camera obstruction or camera failure and generating a fault flag while allowing the audiometric test to continue in a degraded mode reduces unnecessary test interruptions and rework, improving overall test throughput and reliability compared to systems that halt or require re-imaging when visual tracking is unavailable.

[0084] In an optional embodiment, when the imaging data indicates camera obstruction, camera failure, or insufficient image quality for reliable placement detection, the system is configured to generate a fault flag and continue the audiometric test in a degraded operational state without performing image-based placement correction. Detecting camera obstruction or camera failure and generating a fault flag while allowing the audiometric test to continue in a degraded mode reduces unnecessary test interruptions and rework, improving overall test throughput and reliability compared to systems that halt or require re-imaging when visual tracking is unavailable.

[0085] Optionally, the method comprises

[0086] detecting the marker that is coloured;

[0087] determining that the coloured marker is not visible due to environmental conditions; and

[0088] selecting at least one of:

[0089] (i) a QR. code,

[0090] (ii) an augmented reality marker, or(iii) an invisible infra red -reflective pattern, to perform the detection of the placement of the headphone.

[0091] The selection of the alternative marker may be predetermined or dynamically selected based on detected environmental parameters or system confidence levels. This enables continued and accurate determination of headphone placement when a primary marker is not reliably detectable, thereby maintaining robustness of the audiometric test under adverse conditions. In addition, this enhances placementdetection robustness under adverse lighting / occlusion, reduces re-capture / repositioning cycles, and improves correction latency and overall test throughput versus fixed single-marker visible-only systems. In scenarios where no alternative marker is reliably detectable, this further enables the system to flag reduced visual confidence and continue the audiometric test in a simplified operational mode, thereby avoiding unnecessary test interruption while maintaining controlled fault handling. In a second aspect, the present disclosure provides a system for automated detection of a placement of a headphone on an ear of a user for an audiometric test, the system comprising:

[0092] a camera configured to obtain imaging data of the headphone coupled to the ear of the user, wherein the headphone comprises a marker;

[0093] the headphone, configured to be used for an audiometric test; an image processing unit configured to:

[0094] -process the imaging data to automatically detect the marker;

[0095] -determine, based on the detected marker, if a first placement of the headphone on the ear of the user corresponds toa predetermined placement of the headphone to perform the audiometric test; and

[0096] -based on the determination, perform at least one of the following:

[0097] -provide a feedback signal to the user, -reconfigure automatically an audio channel of the headphone, or

[0098] -monitor the obtained imaging data during the audiometric test.

[0099] The system determines whether a placement of the headphone corresponds to a predetermined placement suitable for an audiometric test. Based on this determination, the system automatically reconfigures or swaps an audio channel without requiring manual hardware adjustment of the headphone. As a result, the audiometric test may continue without interruption even when physical realignment of the headphone cannot be performed, thereby maintaining audiometric test continuity and improving the accuracy of audiometric data collection. The system provides an integrated hardware and software arrangement that enables real-time monitoring and performs automatic corrective action during the audiometric test, thereby reducing dependence on the operator, minimizing human error, and improving the accuracy of the audiometric test.

[0100] The camera may be configured to obtain imaging data of the headphone while the headphone is coupled to the ear of the user. The camera may be integrated into a computing device used for the audiometric test, such as a smartphone, tablet, laptop, or dedicated audiometric terminal, or may be provided as a separate peripheral device. The camera may capture one or more images or image sequences representing at least aportion of the user's head and the headphone earpieces, including the marker provided on the headphone.

[0101] In an embodiment, the camera operates in the visible spectrum and captures colour imaging data. The camera may be configured to capture infrared or near-infrared imaging data, when the marker comprises IR-reflective or IR-emissive material. The camera may comprise a depth sensor or stereo camera that is capable of capturing depth information. The depth information may help in distinguishing the headphone from background objects and in compensating for head movement during the audiometric test.

[0102] The headphone is configured for the audiometric test and may comprise a left earpiece and a right earpiece. Each earpiece may be provided with at least one marker that is detectable in the imaging data obtained by the camera. The marker may comprise, for example, a colour region, a pattern, a symbol, an alphanumeric character, or a combination thereof, which enables differentiation between the left and right earpieces. In an embodiment, each earpiece is associated with a distinct colour.

[0103] The image processing unit may be implemented using one or more processors, such as a central processing unit (CPU), a graphics processing unit (GPU), a neural processing unit (NPU), or a combination thereof. The image processing unit may apply colour thresholding and segmentation technique when the marker is colour marker. The image processing unit may employ a template matching technique when the marker comprises pre-defined patterns or symbols. The image processing unit may employ a feature extraction and matching technique when the marker or earpiece includes visual features. A combination of the above techniques may be used based on image quality.

[0104] The camera may be communicatively connected to at least one input interface via a network. The network may be implemented as a wirednetwork, a wireless network, or a combination thereof. In an embodiment, the network comprises a public or private data network, including the Internet. The camera may be implemented using any suitable image-capturing device that is capable of acquiring imaging data of the headphone coupled to the user's ear. For example, the camera may be integrated into or associated with a mobile communication device, a tablet computer, a personal digital assistant, a laptop or notebook computer, a desktop computer, or a smartphone. In some embodiments, the camera is integrated into a portable computing device. In an embodiment, the system employs a server that is communicatively coupled to the camera via the network. The server may be implemented as a local computing device, such as a tablet computer, a desktop computer, a personal computer, or an electronic notebook computer. Alternatively, the server may be implemented as a cloud-based computing service accessible via the Internet. The server is configured to perform at least part of the data processing, including segmentation of a portion of the captured imaging data corresponding to the headphone and, where applicable, detection of one or more markers associated with the headphone.

[0105] In another embodiment, the system is implemented as a portable computing device comprising an integrated camera for capturing imaging data of the headphone coupled to the ear of the user. The portable computing device may further be configured to capture imaging data of at least part of the user's face to determine the headphone placement relative to the user's ear. The portable computing device may comprise a smartphone, a tablet, a web pad, or a laptop computer equipped with a camera.

[0106] Depending on the implementation, the portable computing device may communicate captured imaging data to a remote server for processing, or may be configured to perform all or part of the required imageprocessing locally. Local processing may be advantageous, for example, in scenarios requiring low latency, offline operation, or real-time monitoring during the audiometric test.

[0107] As described above, the imaging data of the headphone may be captured using any suitable image-capturing device. The imaging data may be stored and processed in one or more standard image formats, including, by way of example, JPEG, PNG, GIF, TIFF, PSD, PDF, EPS, or other formats suitable for automated image processing and analysis.

[0108] The camera and the image processing unit may communicate through a local interface, such as an application programming interface (API) endpoint, provided by the system. In operation, the camera is configured to transmit data indicative of the placement of the headphone to the image processing unit through the local interface. The image processing unit may be configured to periodically query the API endpoint for updated placement information or, alternatively, to receive event-driven notifications, for example, through WebSocket-based push mechanisms, indicating a change in detected placement that may require corrective action.

[0109] In an embodiment, the camera transmits data associated with the placement of the headphone to the image processing unit through a shared memory region or a driver-level interface. Such interfaces may be implemented, for example, as a virtual device node or a kernel-managed buffer operated by the system. This arrangement enables low-latency data transfer and helps with real-time analysis of the imaging data during the audiometric test.

[0110] Data related to the placement of the headphone may be exchanged using one or more standard communication protocols, including, but not limited to, TCP / IP sockets, Universal Serial Bus (USB) Human Interface Device (HID) reports, or Bluetooth Low Energy (BLE) Generic Attribute Profile(GATT) characteristics. The selection of a particular communication mechanism may depend on the hardware configuration of the system, such as whether the camera is integrated within the same device as the image processing unit or is provided as an external device.

[0111] Optionally, the image processing unit is further configured to automatically reconfigure the audio channel at least one of (i) reconfigure a first audio channel intended for a left earpiece to a right earpiece of the headphone positioned on a left ear, or (ii) reconfigure a second audio channel intended for the right earpiece to the left earpiece of the headphone positioned on a right ear.

[0112] The audio channel for the headphone refers to a signal path configured to provide an audio signal to the earpiece of the headphone. Each audio channel is operable to transmit sound corresponding to the left or right audio channel. The reconfiguration of the audio channel of the headphone compensates for the reversed / incorrect placement of the headphone, thereby enabling the audiometric test to proceed without requiring physical realignment of the headphone.

[0113] Optionally, the headphone comprises a plurality of sensors configured to provide an additional signal to detect the placement of the headphone on the ear of the user.

[0114] The additional signal obtained from the plurality of sensors are used with imaging data to determine whether the headphone is correctly positioned for the audiometric test. The plurality of sensors enables multi-modal placement detection of the headphone, thereby improving reliability, and accuracy of headphone placement determination, particularly under conditions where image-based detection alone may be degraded, such as poor illumination, partial occlusion, or user movement.The plurality of sensors may be mounted on the headphone. The plurality of sensors may be as EEG electrodes, hyperspectral detectors, scales, Micro-Electro-Mechanical Systems (MEMS) sensors, light-emitting diodes, infrared (IR) emitters, or other sensor modalities. The plurality of sensors may also be mounted on a stand or a user's chair.

[0115] In a third aspect, the present disclosure provides a computer-readable storage medium having instructions stored thereon which, when executed by a processor, cause the processor to perform a method according to the above method.

[0116] The advantages of the present system, the present method, and the associated computer-readable storage medium are thus disclosed above, wherein automated capture and processing of the imaging data enable accurate detection of headphone placement and enable real-time monitoring during the audiometric test. Automatic analysis of the imaging data provides timely feedback and corrective action, such as audio channel reconfiguration, thereby maintaining correct test conditions, reducing interruptions, and improving the accuracy and consistency of audiometric measurement.

[0117] DETAILED DESCRIPTION OF THE DRAWINGS

[0118] FIG. 1 is a schematic illustration of a system 100 for automated detection of a placement of a headphone 108 on an ear of a user 110 for an audiometric test in accordance with an embodiment of the present disclosure. The system 100 includes a camera 106, an image processing unit 102, and the headphone 108. The headphone 108 is configured to be used for an audiometric test. The camera 106 is configured to obtain imaging data of the headphone 108 coupled to an ear of the user 110 within a field of view 118 of the camera 106. In an embodiment, the camera 106 is oriented at an angle such that both the headphone 108 and the user 110 are captured within a common field of view. The camera106 may oriented at a close-up configuration, a mid-framing configuration and a wider environmental framing configuration, relative to the camera 106 during the audiometric test. In an embodiment, the camera 106 employs the field of view 118 of approximately 50 to 90 degrees, enabling reliable monitoring of the headphone 108 relative to the user 110.

[0119] The headphone 108 includes a first marker 114A on a left earpiece 112B and a second marker 114B on a right earpiece 112A. The markers (114A, 114B) may include a colour. For example, the left earpiece 112B includes a red marker, and the right earpiece 112A includes a blue marker, or vice versa. The image processing unit 102 is configured to process the imaging data to automatically detect the marker (114A, 114B). The image processing unit 102 processes the imaging data using image processing. The image processing unit 102 determines if a first placement of the headphone 108 on the ear of the user 110 corresponds to a predetermined placement of the headphone 108, based on the detected marker. The image processing unit 102 provides a feedback signal to the user 110, reconfigures an audio channel of the headphone 108, or monitors the obtained imaging data during the audiometric test.

[0120] In an embodiment, the camera 106 is operatively coupled to an image processing unit 102 through an internal high-speed interface. The camera 106 is configured to capture imaging data representing the headphone 108 positioned on the user 110. The headphone 108 may be communicatively connected to the image processing unit 102 through a network 104 for performing the audiometric test. The network 104 may be implemented as a wired connection, for example via an audio jack or a universal serial bus (USB) connector through a headphone cable 116, or as a wireless connection, for example using Bluetooth® or Wi-Fi communication.The headphone 108 may comprise at least one electro-acoustic driver configured to convert an electrical audio signal into audible sound. Each driver may include a diaphragm, a voice coil, and a magnetic structure, wherein application of the electrical audio signal causes vibration of the diaphragm to generate sound waves. The at least one driver may be housed within an earpiece, such as an earcup or ear-shell, which provides structural support and influences acoustic performance.

[0121] The earpieces may be implemented in an open-back, closed-back, or semi-open configuration, depending on acoustic and isolation requirements. Each earpiece may further include an ear pad or cushion configured to provide user comfort, facilitate sealing around the ear, and improve sound quality during use. The headphone 108 may further comprise a headband connecting the earpieces and configured to secure the headphone on the head of the user 110. The headband may be adjustable and padded. In alternative embodiments, the headphone 108 may be implemented as an in-ear or earbud-type device, where compact housings are configured to fit within the ear canal or the outer ear. Electrical interconnections, such as internal wiring and connectors, may be provided to transmit electrical audio signals from an external source to the at least one driver.

[0122] Prior to applying any corrective action, the image processing unit 102 may be configured to detect and record anomalies or warning events indicative of incorrect or reversed placement of the headphone 108. Such events are logged to enable identification of improper headphone placement during the audiometric test. Once the system 100 automatically reconfigures the audio channels or provides a feedback signal prompting the user 110 to adjust the headphone 108, imaging data and audiometric data acquired after correction may be tagged as valid, thereby distinguishing reliable test data from data collected under incorrect test conditions.In some embodiments, each detected orientation anomaly and the corresponding corrective action are stored in a time-stamped event log. The event log may include, for example, an anonymised user or patient identifier, a test session identifier, and an indication of whether the corrective action was performed automatically or following user intervention. Such structured logging supports traceability of test conditions, facilitates clinical record-keeping, and enables subsequent auditing or quality control.

[0123] In one embodiment, physical characteristics of the headphone 108, such as shape, geometry, form factor, or asymmetric structural features, are used by the system 100 to determine whether a detected earpiece corresponds to a left earpiece 112B or a right earpiece 112A. In another embodiment, the headphone cable 116 serves as an identifiable physical landmark providing a positional reference relative to the ear and head of the user 110.

[0124] The system 100 is configured to detect the headphone cable 116 extending from an earpiece of the headphone 108. Based on a predefined expected position of the headphone cable 116 relative to anatomical features of the user's head, the image processing unit 102 determines whether the detected earpiece corresponds to the left earpiece 112B or the right earpiece 112A. If the detected position of the headphone cable 116 deviates from the expected position, the system 100 determines that the headphone 108 is incorrectly oriented and initiates a corresponding corrective action.

[0125] FIG. 2 is an illustration of a system 200 for automated detection of a placement of a headphone 216 on an ear of a user 218 for an audiometric test in accordance with an embodiment of the present disclosure. The system 200 includes a camera 202, an image processing unit 220, and the headphone 216 associated with the user 218. Theimage processing unit 220 includes an indicator detection module 206, a first condition detection module 208, a pause module 210, a second condition identification module 212, and a resume module 214. The camera 202 captures and obtains facial image data of a user 218 along with the headphone 216 for an audiometric test. The indicator detection module 206 detects one or more indicators associated with the user 218 by processing the facial image data using a facial analysis. The first condition detection module 208 identifies a first condition when the one or more indicators that are detected persist during the audiometric test.

[0126] The pause module 210 automatically pauses the audiometric test in response to the identified first condition (e.g., a state of the user 218 indicative of discomfort, fatigue, stress, reduced attentiveness) and generates a notification to at least one of an operator or the user 218. The second condition identification module 212 identifies a second condition (e.g. a relaxed or stable state of the user 218) when the one or more indicators that are detected persist. The resume module 214 resumes the audiometric test upon determining that the user 218 has returned to the second condition. The image processing unit 220 is calibrated based on a movement of the user 218. For example, the movement of the user 218, such as head tilting or turning.

[0127] Referring next to FIG. 3, there is shown an example illustration of a dock 300 that supports a relative positioning of components during an audiometric test, representing a system in accordance with an embodiment of the present disclosure.

[0128] The dock 300 is configured to support a headphone 302 and a user device 304 in a pre-defined spatial arrangement. The headphone 302 is removably mounted on the dock 300 in a pre-defined orientation that is maintained during the audiometric test. The dock 300 thus defines a stable reference position for the headphone 302 relative to the userdevice 304. The user device 304 comprises a camera 306, which is oriented such that its field of view 310 encompasses at least the headphone 302 and, during use, a portion of a user (not shown). By mounting both the headphone 302 and the user device 304 on the dock 300, a consistent, unobstructed, and repeatable field of view is provided to the camera 306. This configuration enables reliable acquisition of imaging data for monitoring a placement and orientation of the headphone 302 relative to the user during the audiometric test.

[0129] In some embodiments, the camera 306 is further configured to capture facial features of the user during the audiometric test, thereby enabling image-based analysis of the headphone placement and, optionally, monitoring of user-related indicators. This arrangement, as shown in figure, provides a controlled imaging environment that reduces variability due to changes in the relative positioning of the camera 306 and the headphone 302.

[0130] In some embodiments, the camera 306 is configured for capturing a facial image of the user during the audiometric test. In this configuration, the camera 306 is oriented at an angle directed toward a facial region of the user, such that the user's face fills most of a captured frame. The camera 306 corresponds to a close-up configuration (i.e., head-and-shoulders), in which close-ups provide higher-definition imaging of the user. In this configuration, the camera 306 may employ the field of view 310 of approximately 50 to 60 degrees.

[0131] In an embodiment, the camera 306 corresponds to a mid-framing configuration during the audiometric test. In this configuration, the camera 306 employs the field of view 310 of approximately 60 to 75 degrees, to capture a head-to-chest region with partial shoulders and background, of the user.In an embodiment, when the headphone 302 is communicatively connected to the camera 306 through a wireless network for performing the audiometric test, the camera 306 is operated in a wider environmental framing configuration. In this configuration, the camera 306 is oriented at a wider angle relative to the dock 300 and employs the field of view 310 of approximately 75 to 90 degrees to capture the facial image of the user and optionally, one or more additional individuals present in an environment of the dock 300, such as an operator. This arrangement enables remote monitoring or assisted testing setups by allowing visual supervision of the audiometric test without interfering with the wireless operation of the headphone 302.

[0132] In one embodiment, the user device 304 may comprise a smartphone, a tablet device, or a laptop computer including an integrated camera 306. The user device 304 further includes a display 308. The display 308 is configured to present a visual notification to the user via a user interface in response to a feedback signal generated by an image processing unit, for example, to indicate that the placement of the headphone 302 should be corrected. In this manner, the user device 304 provides both image acquisition and user feedback functions during the audiometric test.

[0133] FIG. 4 is a schematic illustration of a system 400 for automated detection of a placement of a headphone 412 on an ear of a user 414 for use in an audiometric test, in accordance with an embodiment of the present disclosure. The system 400 comprises the headphone 412 coupled to the ear of the user 414, an image processing unit 406, and a network 408 providing communication between system components. Imaging data of the headphone 412 coupled to the ear of the user 414 is obtained by a camera (not shown).

[0134] The headphone 412 comprises a right earpiece 410A and a left earpiece 410B, each earpiece including a marker configured to be detected in theimaging data. The headphone 412 is configured for use in an audiometric test, including the delivery of test signals to the respective ears of the user 414.

[0135] The image processing unit 406 comprises a feature identification module 402 and a first placement determination module 404, and is configured to process the imaging data obtained from the camera. In some operating conditions, the marker provided on the headphone 412 may be partially visible or obscured in the imaging data, for example, due to hair, a hand of the user 414, head orientation, or a viewing angle of the camera.

[0136] Where the marker is at least partially obscured in the imaging data, the feature identification module 402 is configured to identify one or more visible features of the headphone 412 and / or the user 414 using image processing techniques. The identified visible features may include, for example, a shape or outline of the headphone 412, edges or contours of the earpieces (410A, 410B) surface patterns, or a spatial relationship between the headphone 412 and anatomical features of the ear of the user 414. Feature identification may be performed using feature-based image processing techniques configured to extract visual features, and / or using a convolutional neural network trained to recognize correct and incorrect placements of the headphone 412, including placements in which one or more markers are partially occluded.

[0137] Based on the detected marker and / or the identified visible features, the first placement determination module 404 determines whether a placement of the headphone 412 on the ear of the user 414 corresponds to a predetermined placement suitable for performing the audiometric test. The determination may include identifying incorrect positioning or reversed placement of the left earpiece 410A and the right earpiece 410B.Based on the placement determination, the image processing unit 406 is configured to perform one or more actions, including providing a feedback signal to the user 414, automatically reconfiguring or swapping an audio channel, i.e. the first audio channel and the second audio channel, of the headphone 412, and monitoring the imaging data during the audiometric test via the network 408. As a result, correct test conditions may be maintained throughout the audiometric test, even if the user 414 moves or if the visibility of the marker changes during operation.

[0138] FIG. 5 is a schematic illustration of an image processing unit 500 of a system in accordance with an embodiment of the present disclosure. The image processing unit 500 comprises a detection module 502, a determination module 504, a switch module 506, an adjustment module 508, and a prompt generation module 510. The image processing unit 500 is configured to receive imaging data of a headphone coupled to an ear of a user. The imaging data may be captured by a camera.

[0139] The detection module 502 is configured to analyse the imaging data in order to determine one or more lighting conditions affecting image quality. The detected lighting condition may include, for example, an overall illumination level, spatial uniformity of illumination, contrast characteristics indicative of shadows or partial occlusions, a type of illumination source, and image noise arising from insufficient illumination or increased sensor gain.

[0140] The determination module 504 is configured to compare the detected lighting condition with a pre-defined threshold lighting condition and to determine whether the detected lighting condition is sufficient for reliable image-based detection of the headphone or associated markers. Based on this determination, the system enables one or more of the switch module 506, the adjustment module 508, and the prompt generation module 510.The switch module 506 is configured to automatically switch from visible-light-based image analysis to infrared-based tracking when the detected lighting condition falls below the pre-defined threshold lighting condition. In some embodiments, the infrared-based tracking is performed using an infrared camera, thereby enabling continued monitoring of a headphone placement under low-light or adverse illumination conditions.

[0141] The adjustment module 508 is configured to automatically modify one or more imaging parameters of the camera in response to the detected lighting condition. Such imaging parameters may include, for example, exposure time, sensor gain, white balance, or focus settings, in order to improve image quality and maintain reliable detection performance.

[0142] The prompt generation module 510 is configured to generate a prompt for an operator or user when the detected lighting condition is insufficient for reliable image analysis. The prompt may indicate that the lighting condition should be adjusted and may be presented as a visual notification, such as a message displayed on a screen, and / or as an audio notification, thereby facilitating timely corrective action.

[0143] FIG. 6 is a schematic illustration of an image processing unit 600 of a system in accordance with an embodiment of the present disclosure. The image processing unit 600 includes a detection module 602, a determination module 604, and a selection module 606. The detection module 602 is configured to detect a marker provided on a headphone coupled to an ear of a user from imaging data captured during an audiometric test. In one example, the marker is a coloured marker detectable in visible-light imaging data. The determination module 604 is configured to evaluate the imaging data to determine whether the coloured marker is reliably visible. Such a determination may be based, for example, on detected image quality parameters, illumination conditions, occlusion, or contrast levels.When the determination module 604 determines that visibility of the coloured marker is insufficient, for example, due to adverse environmental conditions, the selection module 606 is configured to dynamically select an alternative marker type to enable continued and accurate detection of a headphone placement. The alternative marker type may include, by way of example, (i) a Quick Response (QR) code, which enables improved detection based on pattern recognition; (ii) an Augmented Reality (AR) marker, which allows estimation of pose and orientation of the headphone relative to the user's ear; or (iii) an infrared-reflective or infrared-encoded pattern, which is detectable using an infrared imaging mode.

[0144] By adaptively switching between different marker types in response to detected imaging conditions, the image processing unit 600 maintains reliable detection of headphone placement across varying environmental conditions, thereby improving the accuracy of the placement detection during the audiometric test.

[0145] FIG. 7 is an illustration of an architecture of a system 700 showing an interaction between software components 711 and hardware components 709 for an audiometric test in accordance with an embodiment of the present disclosure.

[0146] As shown, the system 700 includes the hardware components 709, which include a camera 701, a headphone 704 with markers, and a display 710. The camera 701 captures imaging data 702 as a visual input 705 representing the headphone 704 with markers positioned on a user and provides it to an image processing unit 703. The markers disposed on the headphone 704 enable reliable detection of a placement of the headphone 704 within the captured imaging data 702. The software components 711 include the image processing unit 703 and an audiometric testing system 707. The image processing unit 703 receivesand processes the imaging data 702 to extract orientation data 706 corresponding to the placement and orientation of the headphone 704 relative to the user's ears.

[0147] The image processing unit 703 may further analyze physiological data 708, such as facial expressions or head movements, derived from the imaging data 702. The audiometric testing system 707 receives the orientation data 706 and the physiological data 708 and determines whether the detected headphone placement satisfies a pre-defined placement of the headphone 704. If the determination indicates reversed placement of the headphone 704, the audiometric testing system 707 enables channel adjustment 713, which is applied to the headphone 704 to automatically correct audio channel assignment of the headphone 704 without requiring physical repositioning of the headphone 704. The audiometric testing system 707 may also generate a feedback 712 if the determination indicates the reversed placement of the headphone 704. The feedback 712 is provided to the display 710 to visually instruct the user of placement status or testing progress. If the determination indicates correct placement, the camera 701 continues to monitor the provided imaging data 702 during the audiometric test.

[0148] FIG. 8 illustrates a flowchart of a method for automatically detecting and correcting headphone placement during an audiometric test in accordance with an embodiment of the present disclosure. As shown in FIG. 8, the method begins at step 801, where the audiometric test session is initiated. At step 802, a camera and a headphone are set up in association with a user. The camera is positioned to capture imaging data of the headphone when worn on the user's ears. At step 803, a calibration system performs a calibration procedure, which may include configuring parameters such as exposure, focus, white balance, and frame rate of the camera.Following the calibration procedure, the method proceeds to step 804, where imaging data of the headphone is continuously or periodically captured during the audiometric test. The captured imaging data is provided to an image processing pipeline at step 805, where a placement of the headphone is analyzed by detecting markers of the headphone in the imaging data, determining their relationship relative to the user's ears and logging findings by tagging the imaging data as invalid or valid. The markers may be a color or a pattern.

[0149] Based on the analysis, the method determines whether the headphone placement is correct. If the placement is determined to be incorrect at step 808, the method proceeds to step 809, to detect a type of error in the headphone placement. The type of error may include, for example, reversed left-right orientation, improper seating of one or both earcups, partial displacement, or misalignment relative to an ear canal. At step 810, an appropriate corrective action is selected based on the detected error type.

[0150] The corrective action comprises a physical fix at step 813 and a softwarebased fix at step 811. For the software-based fix, the method comprises automatically reconfiguring audio channels of the headphone for a detected left-right reversal placement of the headphone at step 812, without requiring any physical repositioning of the headphone. For the physical-based fix, the method comprises notifying the user or an operator at step 814 through a visual, an audio, or a textual prompt to adjust the placement of the headphone.

[0151] After the corrective action is applied, the method returns to step 804 to re-capture imaging data and re-analyze the headphone placement, thereby forming a closed-loop verification process. If the placement of the headphone is determined to be correct at step 806, the method comprises continuous monitoring of the placement of the headphoneduring the audiometric test at step 807, without interrupting test execution. At step 815, it is checked whether the audiometric test has been completed. If the test is not complete (step 816 - No), the monitoring and correction process continues. If the test is complete (step 817 - Yes), the process terminates at step 818.

[0152] At the step 810, the selection of the corrective action may be followed by a verification step ("Adjustment made?"), which serves as a confirmation step to ensure that the physical-based fix, at the step 813, has been completed before returning to capturing the imaging data at the step 804. If the verification step is excluded, the notification is displayed to the user or the operator until the required physical-based fix is detected or confirmed.

[0153] FIG. 9 illustrates a flow diagram of a method for automated detection of a placement of a headphone and indicators of a user for an audiometric test, in accordance with an embodiment of the present disclosure. At step 901, imaging data of a headphone coupled to an ear of the user is captured using a camera. The captured imaging data is received at step 902 as raw image data. At step 903, the raw image data is processed using one or more image pre-processing techniques, including filtering, segmentation, and feature enhancement, to generate processed image data at step 904.

[0154] At step 905, a marker detection operation is performed on the processed image data to detect a marker associated with the headphone. The marker may include a color-coded marker, a pattern, or a machine-readable marker. At step 906, a detected image is generated indicating the location and visibility of the detected marker within the imaging data.

[0155] At step 907, an orientation analysis is performed based on the detected marker. The method checks for a left-side marker on a left earpiece of the headphone at step 909 and for a right-side marker on a right earpieceof the headphone at step 910. Based on the presence, absence, or relative position of the left and right markers, the placement and the orientation of the headphone are determined at step 911, including whether the headphone is correctly placed or reversed. The determined placement is output as orientation data at step 908.

[0156] Physiological analysis is performed on facial image data to detect indicators of the user. The facial image data is captured using the camera. The detected indicators include physiological data at step 912, facial expression data at step 913, pupil dilation data at step 914, and blink rate data at step 915. The detected indicators are physiological data. In parallel, the orientation data is received at step 917.

[0157] At step 916, the user state data is determined based on the orientation data, and the physiological data. At the step 917, audiometric test results are analyzed in view of the determined user state data to identify inconsistencies, artifacts, or confidence levels associated with the audiometric test results. At step 918, analysis results are generated, and at step 919, a decision-making operation is performed based on the analysis results. The decision-making operation may determine whether corrective action is required. At step 920, one or more actions are taken, including automatically correcting audio channel assignment, adjusting test parameters, pausing or resuming the audiometric test, or generating feedback, without requiring physical repositioning of the headphone.

[0158] Referring next to FIG. 10, there is shown an illustration of a flowchart illustrating steps of a method for (namely, a method of) automated detection of a placement of a headphone on an ear of a user for an audiometric test in accordance with an embodiment of the present disclosure. At step 1002, imaging data of the headphone coupled to the ear of the user is obtained. The headphone includes a marker. At step 1004, the imaging data is processed using an image processing toautomatically detect the marker. At step 1006, the detected marker is checked to determine whether a first placement of the headphone on the user's ear aligns with a predetermined placement of the headphone to perform the audiometric test. At step 1008, based on the determination, at least one of: a feedback signal is provided to the user, an audio channel of the headphone is reconfigured automatically, or the obtained imaging data during the audiometric test is monitored, is performed.

[0159] In FIG. 11, there is shown an illustration of a block diagram of a system, including a computing architecture, in accordance with an embodiment of the present disclosure. The exploded view depicts a system that comprises at least one input interface 1102, a control module that comprises a data processing arrangement 1104, a memory 1106 and a non-volatile storage 1108, processing instructions 1110, a shared / distributed storage 1112, and an image capturing device that comprises a processor 1114, a memory 1116 and a non-volatile storage 1118 and an output interface 1120. The functions of the data processing arrangement 1104, the least one input interface 1102, are as described above and cooperate together to implement methods of the disclosure as described in the foregoing.

[0160] Modifications to embodiments of the present disclosure described in the foregoing are possible without departing from the scope of the present disclosure as defined by the accompanying claims. Expressions such as "including", "comprising", "incorporating", "have", "is" used to describe and claim the present disclosure are intended to be construed in a nonexclusive manner, namely allowing for items, components or elements not explicitly described also to be present. Reference to the singular is also to be construed to relate to the plural.

Claims

CLAIMS1. A computer-implemented method for automated detection of a placement of a headphone (108, 216, 302, 412) on an ear of a user (110, 218, 414) for an audiometric test, the method comprising:obtaining imaging data of the headphone (108, 216, 302, 412) coupled to the ear of the user (110, 218, 414), wherein the headphone (108, 216, 302, 412) comprises a marker (114A, 114B);processing, using an image processing, the imaging data to automatically detect the marker (114A, 114B);determining, based on the detected marker (114A, 114B), if a first placement of the headphone (108, 216, 302, 412) on the ear of the user (110, 218, 414) corresponds to a predetermined placement of the headphone (108, 216, 302, 412) to perform the audiometric test; and based on the determination, performing at least one of the following:- providing a feedback signal to the user (110, 218, 414), -reconfiguring automatically an audio channel of the headphone (108, 216, 302, 412), or-monitoring the obtained imaging data during the audiometric test.

2. The method according to claim 1, wherein the automatic reconfiguration of the audio channel comprises at least one of (i) reconfiguring a first audio channel intended for a left earpiece (112B, 410B) to a right earpiece (112A, 410A) of the headphone (108, 216, 302, 412) positioned on a left ear, or (ii) reconfiguring a second audio channel intended for the right earpiece (112A, 410A) to the left earpiece (112B, 410B) of the headphone (108, 216, 302, 412) positioned on a right ear.

3. The method according to claim 1 or 2, wherein the method further comprisesobtaining facial image data of the user (110, 218, 414) that is captured using an image capturing unit (106, 202, 306) for the audiometric test;detecting one or more indicators of the user (110, 218, 414) by processing the facial image data using a facial analysis;identifying a first condition when the one or more indicators that are detected persist during the audiometric test;automatically pausing the audiometric test in response to the identified first condition and generating a notification to at least one of an operator or the user (110, 218, 414);identifying a second condition when the one or more indicators that are detected persist during the audiometric test; andresuming the audiometric test upon determining that the user (110, 218, 414) has returned to the second condition.

4. The method according to claim 2, wherein the marker (114A, 114B) comprises a pattern, a colour, an alphanumeric code, or a symbol, indicating whether the earpiece is the left earpiece (112B, 410B) or the right earpiece (112A, 410A).

5. The method according to claim 3, wherein the method comprises positioning the image capturing unit (106, 202, 306) at a distance of between 0.5 meter (m) and 3 meter (m) in front of the user (110, 218, 414) and at an angle between 30° and 75° relative to a frontal plane of the user (110, 218, 414) so that the earpiece is visible in the imaging data.

6. The method according to claim 3 or 5, wherein the image capturing unit (106, 202, 306) captures the imaging data at a frame rate of 10 to 30 frames per second.

7. The method according to any of the preceding claims, wherein the image processing is performed using a machine learning model that is trained with at least one of:(i) a dataset of images with correctly and incorrectly placed headphones, markers, subtle facial landmarks, head pose, skin tone, and headphone material;(ii) an augmented dataset that comprises images in low-light conditions, stand, user's chair, and additional data from one or more sensors of the headphone; or(iii) a dataset that comprises different headphone textures or images of users with different demographic characteristics.

8. The method according to claim 1 or 3, wherein the imaging data of the headphone (108, 216, 302, 412) or the facial image data of the user (110, 218, 414) is obtained using at least one of: an infrared camera, a depth sensor, a hyperspectral detector or a thermal camera for detecting the placement of the headphone (108, 216, 302, 412) under varying lighting conditions, or partial occlusion.

9. The method according to any of the preceding claims, wherein when the marker (114A, 114B) of the headphone (108, 216, 302, 412) is at least partially blocked from view in the imaging data of the headphone (108, 216, 302, 412), the method comprisesidentifying one or more visible features in the imaging data using a feature-based image processing or a convolutional neural network; and determining, based on the one or more identified visible features, if the first placement of the headphone (108, 216, 302, 412) on the ear of the user (110, 218, 414) corresponds to the predetermined placement of the headphone (108, 216, 302, 412) to perform the audiometric test.

10. The method according to claim 1 or 3, wherein the method comprisesdetecting a lighting condition within the imaging data of the headphone (108, 216, 302, 412) or the facial image data of the user (110, 218, 414);determining whether the lighting condition is below a pre-defined threshold lighting condition; andbased on the determination, performing at least one of the following:switching automatically from the detected lighting condition to infrared-based tracking, wherein the infrared-based tracking is performed using an infrared camera,adjusting automatically one or more imaging parameters of the image capturing unit (106, 202, 306), based on the pre-defined threshold lighting condition, orgenerating a prompt to the operator to adjust the lighting condition.

11. The method according to any of the preceding claims, wherein the method comprisesdetecting the marker (114A, 114B) that is coloured; determining that the coloured marker (114A, 114B) is not visible due to environmental conditions; andselecting at least one of:(i) a QR. code,(ii) an augmented reality marker, or(iii) an invisible infra red -reflective pattern, to perform the detection of the placement of the headphone (108, 216, 302, 412).

12. A system (100, 200, 300, 400) for automated detection of a placement of a headphone (108, 216, 302, 412) on an ear of a user (110, 218, 414) for an audiometric test, the system (100, 200, 300, 400) comprising:a camera (106, 202, 306) configured to obtain imaging data of the headphone (108, 216, 302, 412) coupled to the ear of the user (110, 218, 414), wherein the headphone (108, 216, 302, 412) comprises a marker (114A, 114B);the headphone (108, 216, 302, 412), configured to be used for an audiometric test;an image processing unit (102, 220, 406, 500, 600, 703) configured to:- process the imaging data to automatically detect the marker (114A, 114B);- determine, based on the detected marker (114A, 114B), if a first placement of the headphone (108, 216, 302, 412) on the ear of the user (110, 218, 414) corresponds to a predetermined placement of the headphone (108, 216, 302, 412) to perform the audiometric test; and- based on the determination, perform at least one of the following:- provide a feedback signal to the user (110, 218, 414), -reconfigure automatically an audio channel of the headphone (108, 216, 302, 412), or- monitor the obtained imaging data during the audiometric test.

13. The system (100, 200, 300, 400) according to claim 12, wherein the image processing unit (102, 220, 406, 500, 600, 703) is further configured to automatically reconfigure the audio channel at least one of (i) reconfigure a first audio channel intended for a left earpiece (112B, 410B) to a right earpiece (112A, 410A) of the headphone (108, 216, 302, 412) positioned on a left ear, or (ii) reconfigure a second audio channel intended for the right earpiece (112A, 410A) to the left earpiece (112B,14. The system (100, 200, 300, 400) according to claim 12 or 13, wherein the headphone (108, 216, 302, 412) comprises a plurality of sensors configured to provide an additional signal to detect the placement of the headphone (108, 216, 302, 412) on the ear of the user (110, 218, 414).

15. A computer-readable storage medium having instructions stored thereon which, when executed by a processor, cause the processor to perform a method according to any of method claims 1 - 11.

Images