Wearable device comprising hearing aid module configured to operate with low-power consumption and a method for operating said wearable device
A context-aware method and device for hearing aids automatically manage power consumption by activating modules based on listening environments, addressing the need for manual activation and extending battery life.
Patent Information
- Application Number
- PCT/EP2026/064070
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-14
- Filing Date
- 2026-05-13
- Publication Date
- 2026-11-19
AI Technical Summary
Hearing aid devices require manual activation and deactivation due to high power consumption, leading to rapid battery depletion and inability to remain continuously powered on.
A context-aware method and device that utilizes an Environmental Context Classifier and event-based camera to automatically activate hearing aid modules based on listening environments, minimizing power consumption and extending battery life.
The method and device automatically manage hearing aid functionalities, reducing power consumption and eliminating the need for manual activation, thereby extending battery life and improving user experience.
Smart Images

Figure EP2026064070_19112026_PF_FP_ABST
Abstract
Description
[0001] WEARABLE DEVICE COMPRISING HEARING AID MODULE CONFIGURED TO OPERATE WITH LOW-POWER CONSUMPTION AND A METHOD FOR OPERATING SAID WEARABLE DEVICE
[0002] The present invention relates to a wearable device comprising hearing aid module configured to operate with low-power consumption, and to a method for operating said wearable device .
[0003] In the current state of the art, hearing aid devices are known .
[0004] The hearing aid devices of the prior art can be implemented with an Environmental Context Classifier . The classifier assesses incoming sounds and categorizes audio data into various listening environments, enabling an adjustment of hearing aid functionalities, comprising modifying amplification levels or implementing noise reduction algorithms . The prior art classifier involves only the audio data (plus Inertial Movement Unit IMU in some cases) .
[0005] Beamforming and similar advanced features of audio processing methods in prior art devices like smart eyewear and hearing aid devices usually employ technologies like Field Programmable Gate Arrays (FPGA) or System on Chip (SoC) . These prior art devices must meet strict latency requirements to ensure they are usable in real-time way, i . e . within a few milliseconds, for sound amplification . This requirement leads to high power consumption which prevents advanced features from being always switched on during use . Additionally, certain situations may arise in which users may not require or may prefer that their hearing aid devices remain deactivated. While users can manually switchthese advanced features on and off, this process can be cumbersome .
[0006] A drawback of the prior art is that the user of the wearable device is required to manually determine whether to activate the hearing aid device . The hearing aid device cannot remain continuously powered on due to their high power consumption, which results in the rapid depletion of the device ' s battery .
[0007] The hearing aid devices are activated by the user only in specific scenarios, such as during conversations or when interacting with others, while remaining deactivated in other scenarios to mitigate excessive power consumption and prevent rapid depletion of the device ' s battery .
[0008] An object of the present invention is to provide a method for operating a wearable device, activating a hearing aid module in a context-aware manner, extending the battery life of the device, minimizing power consumption, and eliminating a need for manual activation or deactivation of the hearing aid module by the user .
[0009] According to the invention, this object is achieved by a method for operating a wearable device according to claim 1 .
[0010] Another object of the present invention is to provide a wearable device activating a hearing aid module in a context-aware manner, extending the battery life of the device, minimizing power consumption, and eliminating a need for manual activation or deactivation of the hearing aid module by the user .
[0011] According to the invention, this other object is achieved by a wearable device according to claim 9.A further object of the present invention is to provide head-mounted device comprising a wearable device activating a hearing aid module in a context-aware manner, extending the battery life of the device, minimizing power consumption, and eliminating a need for manual activation or deactivation of the hearing aid module by the user .
[0012] According to the invention, this further object is achieved by a head-mounted device according to claim 17.
[0013] Other characteristics are provided in the dependent claims .
[0014] The characteristics and the advantages of the present invention will become more evident from the following description, which is exemplary and not limiting, with reference to the appended schematic drawings, in which :
[0015] figure 1 is a perspective view of a head-mounted device, for example eyeglasses, which mounts a wearable device of the present invention, wherein the wearable device comprises microphones, speakers, event-based camera, at least one processor and at least one memory;
[0016] figure 2 is a flow chart of a method for operating the wearable device according to the present invention .
[0017] With reference to the figures mentioned above, a head-mounted device 10, such as for example eyeglasses, is shown in figure 1.
[0018] The head-mounted device 10 mounts a wearable device comprising a multiplicity of microphones 20 spaced apart from each other, two speakers 30, at least one eventbased camera 40, and at least one processor and at least one memory .
[0019] Each microphone 20 of said multiplicity ofmicrophones 20 is adapted to receive a sound which is more generally referred to as an audio signal .
[0020] Said at least one memory contains at least the steps of a method 100, shown in figure 2, for operating the wearable device .
[0021] Said at least one processor implements the steps of the method 100 for operating the wearable device .
[0022] Figure 1 shows the wearable device mounted with the head-mounted device 10.
[0023] Preferably as shown in figure 1 the microphones 20 of the wearable device are mounted in order to be spaced apart from each other on the temples and / or on a nose bridge and / or on the frame of the eyeglasses, or more generally, of the head-mounted device .
[0024] Preferably the wearable device comprises seven microphones 20 disposed spaced apart from each other in order to cover a wider reception angle .
[0025] Preferably said microphones 20 are omnidirectional . Advantageously, the distance between the microphones 20 enables improved detection of the direction from which the received sound originates . The sound direction detection is performed by said at least one processor .
[0026] Advantageously the spacing between microphones 20 is necessary for beamforming, as the time delay it creates allows the beamforming process to constructively combine the audio signals received from the direction of the speech source .
[0027] As shown in figure 1, preferably the wearable device comprises two speakers 30, i . e . one speaker 30 for each ear of a user wearing the wearable device .
[0028] The user is a person who wears the wearable device .It is not necessary that the wearable device comprises two speakers 30 mounted for each of the user ' s ears, and alternatively it is possible to envisage that it is sufficient that the wearable device comprises only one speaker 30 mounted for only one ear of the user .
[0029] Preferably, as shown in figure 1 the wearable device comprises also an event-based camera 40.
[0030] Preferably by using data from the event-based camera 40, said at least one processor identifies a direction from which the speech comes by identifying a face of a person who is speaking.
[0031] For example, the processor identifies the face of the person in the direction of the user, preferably by using a YOLO (You Only Look Once) -based algorithm, which detects if there is a face in the scene of the user, wherein the scene is an angle of view of the user .
[0032] Preferably by using data from the event-based camera 40, the processor detects if there are lips movements on the face of the person in the scene of the user .
[0033] For example, the processor detects the lips movements on the face of the person, preferably by using the YOLO-based algorithm which after detecting that there is a face of a person in the scene of the user, the processor, preferably by using the YOLO-based algorithm, extracts a Region Of Interest (ROI) of the lips . After that operation, the processor operates a CNN (Convolutional Neural Network) , which is a classifier, and the classifier recognizes if the lips are moving and preferably if the person is talking in the scene of the user wearing the wearable device, i . e . speech direction detection .Said at least one memory of the wearable device comprises a hearing aid module .
[0034] Said hearing aid module comprises : speech detection functionality, omnidirectional sound amplification functionality, speech direction detection functionality and directional sound amplification functionality .
[0035] These functionalities are algorithms . These functionalities are implemented by said at least one processor as sub steps of the steps of the method 100.
[0036] Preferably speech detection functionality, speech direction detection functionality of the present invention are implemented for example by the processor with an Environmental Context Classifier, which is a classifier .
[0037] Alternatively, it is possible to envisage that other classifiers can be used by the processor in order to implement the steps of a method 100 for operating the wearable device .
[0038] Advantageously said classifier is operated by said at least one processor .
[0039] Said classifier of the wearable device assesses incoming sounds and categorizes audio data into a multiplicity of listening environments, enabling an adjustment of said speech detection functionality and / or speech direction detection functionality .
[0040] Preferably said classifier of the wearable device assesses incoming data of event-based camera 40 to operate said speech direction detection functionality .
[0041] The data of the event-based camera 40 are for example an identification of the direction of the user wearing the wearable device, an identification of a presence of the face of the person in the scene of theuser and preferably a detection of lips movements on the face of the person in the scene of the user .
[0042] After assessing the data of the event-based camera 40, said classifier categorizes said data of the eventbased camera 40 into a multiplicity of listening environments, enabling the adjustment of said speech detection functionality or said speech direction detection functionality .
[0043] Listening environments are for example speech detection, noise detection, surrounding people talking, user is listening to the radio or to the music, the user is looking at someone else, someone is speaking directly in front of the user, other listening environments according to additional features according for example user preferences .
[0044] Said speech detection functionality detects if there is a speech from an audio signal received. Said at least one processor is adapted to detect speech from the audio signals received by said microphones 20.
[0045] Preferably said at least one processor detects the speech from said audio signal received by the microphones 20 by using the classifier .
[0046] Advantageously the speech detection functionality of the wearable device makes a complete audio signal capture from all directions : preferably omnidirectional microphones 20 capture audio signals from all directions .
[0047] Advantageously the wearable device improves the overall audio quality, resulting in a better environmental awareness, ease in social interactions, immersive experience for Augmented Reality (AR) and Virtual Reality (VR) applications .Preferably the modification of amplification levels of the microphones 20 is obtained by using an omnidirectional sound amplification functionality which allows the processor to pick up sounds from all directions evenly by using the microphones 20 and to amplify the audio signal transmitted by said at least one speaker 30.
[0048] Said omnidirectional sound amplification functionality comprises functionalities such as for example modifying amplification levels of said microphones 20 and / or amplification levels of audio signal transmitted to said at least one speaker 30, implementing noise reduction algorithms in order to transmit a noise-free audio signal to said at least one speaker 30.
[0049] Preferably said at least one processor is adapted to do at least one action comprised in a list comprising: to modify amplification levels of at least one microphone 20 of said multiplicity of microphones 20 and / or to modify amplification levels of the audio transmitted to at least one speaker 30; to implement noise reduction algorithms of at least one speaker 30.
[0050] Said speech direction detection functionality comprises at least one of the functionalities of a list comprising face detection, face direction identification, lips movement detection .
[0051] Said at least one processor is adapted to do at least one action comprised in a list comprising: to detect said face or at least one region of said face from which the speech received comes by using data of said at least one event-based camera 40; to identify a direction of a face from which the speech received comesby using data of said at least one event-based camera 40; to detect lips movement on said face or said at least one region of said face by using data of said at least one event-based camera 40.
[0052] Preferably said at least one processor is adapted to do at least one action comprised in a list comprising: to modify amplification levels of said at least one microphone 20; to modify amplification levels of audio transmitted to at least one speaker 30; to implement noise reduction algorithms of said at least one speaker 30 .
[0053] Preferably said event-based camera 40 is used for identifying the direction of a face from which the speech comes .
[0054] Alternatively the data of at least one event-based camera 40 can be implemented also with data of a Time of Flight (ToF) Camera . Said ToF camera can be used to identify the direction of the face of the person from which the speech comes . The ToF camera provides complementary information as distance and / or depth, to the event-based camera 40. This distance information can be fused with that of the event-based camera 40 to improve the estimation of the direction from which the sound comes, i . e . the data of ToF camera can be used to implement the data of event-based camera 40.
[0055] Event-based camera 40 is used also for face detection, face direction detection and lips movements detection .
[0056] Said directional sound amplification functionality amplifies the audio signal received at particular angles from which the speech comes and then transmits the amplificated audio signal containing speech to said atleast one speaker 30.
[0057] Preferably said directional sound amplification functionality comprises beamforming.
[0058] Said at least one processor is adapted to apply a beamforming to the audio signals captured by said microphones 20.
[0059] Beamforming or spatial filtering is a signal processing technique used for directional signal reception and achieves a spatial selectivity .
[0060] Processor achieved beamforming by amplifying and / or combining the audio signals of the microphones 20 received by said at least one processor in such a way that the audio signals of the microphones 20 at particular angles experience constructive interference while others experience destructive interference . Said particular angles experiencing said constructive interference are towards the direction of a selected audio signal to be beamformed, wherein said selected audio signal to be beamformed is said speech . The beamformed audio signal is then transmitted to said at least one speaker 30.
[0061] Preferably and advantageously the processor applies a beamforming technique by using said multiplicity of microphones 20.
[0062] Advantageously the directional sound amplification used for said wearable device allows a selection of collected signals coming from a specific direction, and / or an attenuation of other signals coming from other directions .
[0063] Advantageously the directional sound amplification used for said wearable device allows that the audio signal transmitted to said at least one speaker 30 enjoysa reduction of background noise, an improvement of speech clarity, a focused listening at a specific direction .
[0064] Preferably the wearable device comprises also at least one movement sensor comprised in a list comprising IMU (Inertial Movement Unit) , accelerometers and / or gyroscope .
[0065] Preferably the wearable device comprises also at least one ambient sensor comprised in a list comprising a temperature sensor, a pressure sensor, a humidity sensor, a light sensor .
[0066] As shown in figure 2, the method 100 for operating the wearable device comprising a step of measuring 1 sound intensity . The step of measuring sound intensity 1 provides that if processor finds that measured sound intensity is higher than a predefined threshold value, for example 40 dB, then the processor activates a speech detection step 2, i . e . said speech detection functionality is activated. Said speech detection step 2 provides that said at least one processor detects speech from the sound received by said microphones 20 by using speech detection functionality .
[0067] If speech detection functionality does not detect any speech, then the speech detection functionality is switched off again . Furthermore if omnidirectional sound amplification functionality has been activated, then the processor deactivates also omnidirectional sound amplification functionality and the method 100 returns to step 1 waiting for a measurement of sound intensity higher than a predefined sound intensity value .
[0068] Alternatively it is possible to envisage that the predefined threshold value is set based on the background noise .If speech detection step 2 detects some speech, then omnidirectional sound amplification step 23 is activated by activating said omnidirectional sound amplification functionality .
[0069] Advantageously the processor maintains active speech detection functionality also during omnidirectional sound amplification step 23.
[0070] Preferably said omnidirectional sound amplification step 23 modifies amplification levels of the microphones 20 and transmits amplificated audio signal to said at least one speaker 30, preferably omnidirectional amplification, and preferably implementing noise reduction algorithms, transmitting a noise-free amplificated audio signal to said at least one speaker 30 .
[0071] After that if speech detection is detected during omnidirectional sound amplification step 23, then the speech direction detection step 3 is activated by said at least one processor .
[0072] Said speech direction detection step 3 provides that said at least one processor activates said speech direction detection functionality .
[0073] Preferably if speech direction detection functionality provides to use said event-based camera 40, then preferably said at least one processor activates in this order : face detection functionality, face direction identification functionality and lips movement detection functionality .
[0074] Preferably if said at least one processor by using the event-based camera 40 identifies that there is a face in the direction from which the speech comes, the face detection functionality has been activated by saidat least one processor and then the lips movement detection functionality has been activated by said at least one processor .
[0075] If there is also lips movement detected by said lips movement detection functionality, then said at least one processor activates directional sound amplification step 33 by activating directional sound amplification functionality .
[0076] It is important to point out that said at least one processor activates step by step the speech direction detection functionality, i . e . not all together at the same time .
[0077] If the speech stops or lips stop moving, then directional sound amplification functionality is switched off 32 and also the speech direction detection functionality is switched off, also omnidirectional sound amplification functionality is switched off and the method 100 returns to step 2, i . e . maintaining only active speech detection functionality .
[0078] Preferably said directional sound amplification step 33 activates directional sound amplification functionality comprising beamforming and this case at least two microphones 20 need to be used by said at least one processor to implement beamforming.
[0079] Preferably the steps 1, 2, 23, 3, 33 of the method 100 are in chronological order in order to be activated step by step during an interval of time and not all together at the same time .
[0080] Alternatively if said at least one processor detects said face or at least one region of said face, or if said at least one processor identifies a direction of said face, then said at least one processor detectslips movement, and if said at least one processor detects lips movement on said face or said at least one region of said face, then said at least one processor activates said directional sound amplification functionality .
[0081] Alternatively it is possible to envisage that during step 3 said at least one processor activates speech direction detection functionalities all together at the same time, i . e . if speech is detected, than the face detection functionality, the face direction identification functionality and said lips movement detection functionality have been activated all together at the same time by said at least one processor .
[0082] More preferably said at least one processor activates one or more of the speech direction detection functionalities using also incremental context recognition by using the Environmental Context Classifier, which involves utilizing low-power, always-on sensors for initial classification (e . g. ambient sensors, movement sensors, microphone sound intensity threshold) .
[0083] Preferably in case preliminary analysis indicates a context where activating omnidirectional sound amplification functionality and / or directional sound amplification functionality could be necessary, then higher-power sensors, such as full microphones 20 array and an event-based camera 40, and computational resources are needed to further refine the classification .
[0084] It is important to enlightening that said at least one processor does not activate all the functionalities at once, but the processor activates said speech detection functionality, said omnidirectional soundamplification functionality, said speech direction detection functionality and said directional sound amplification functionality step by step in a way that can be named as incremental sensors activation or more general as incremental functionalities activation .
[0085] For example, if the user is on a tram, train, or vehicle surrounded by people conversing, the algorithm will activate omnidirectional sound amplification functionality when the user is not facing a person who speaks, i . e . the person who is speaking is not in the scene of the user . Directional sound amplification functionality will only be activated when the user is directly facing the person speaking, thereby specifically amplifying that person ' s voice .
[0086] For example, when the user is cooking and listening to the radio, directional sound amplification functionality should not be activated, as it would amplify the undesirable sounds associated with the cooking activity . Since no person is speaking directly in front of the user, i . e . in the scene of the user, only omnidirectional sound amplification will be activated, along with a noise reduction algorithm, if preferred .
[0087] Alternatively it is possible to envisage that all of said functionalities are implemented with said classifier operated by said at least one processor . Preferably said classifier is adapted to assess the sound received and to categorize audio data from sound received into a multiplicity of listening environments, enabling an adjustment of speech detection functionality and / or speech direction detection functionality .
[0088] Preferably said classifier is adapted to assessincoming data of said at least one event-based camera 40 .
[0089] Preferably said classifier is adapted to categorize said data of the event-based camera 40 into a multiplicity of listening environments, enabling the adjustment of said functionalities .
[0090] Preferably and additionally learning algorithms can be applied to the classifier, preferably to the Environmental Context Classifier, in order to tailor the activation of functionalities according to specific user preferences . In this case the classifier comprises an Artificial Intelligence (Al) module that can be trained by the user in supervised way . The Al module learns in a supervised way when the processor should activate the functionalities depending on the situation .
[0091] Alternatively it is possible to envisage that the wearable device comprises at least two microphones 20. A configuration with only two microphones 20 is sufficient; but in this case, amplification is limited to a single direction .
[0092] Alternatively in order to ensure coverage of all potential directions from which the audio signal may originate, it is preferable that the microphones 20 be configured in a quantity from two to eight and spatially separated from each other . A configuration of two to eight microphones 20 is particularly advantageous for beamforming as directional sound amplification functionality .
[0093] Alternatively it is possible to envisage that the wearable device can comprise only one microphone 20. Preferably the only one microphone 20 can be used when the directional sound amplification step 33 may bedifferent from beamforming.
[0094] Alternatively it is possible to envisage that the wearable device comprises only one microphone 20 comprising at least a multiplicity of receiving lobes directed towards a multiplicity of directions .
[0095] Alternatively to obtain directional sound amplification without beamforming it is possible to envisage that the processor increases the audio intensity of at least one of the microphones 20 which is directed towards the direction of the selected signal to be amplified.
[0096] Alternatively to obtain directional sound amplification without beamforming it is possible to envisage that the processor increases at least one receiving lobe of said multiplicity of receiving lobes, wherein said at least one receiving lobe is directed towards the direction of a selected signal to be beamf ormed .
[0097] More generally, it is possible to envisage that the wearable device comprises a multiplicity of microphones 20, or at least one microphone 20, comprising a multiplicity of receiving lobes directed towards a multiplicity of directions to cover a multiplicity of directions from which an audio signal can come .
[0098] Alternatively said microphones 20 may be directional . In the case of directional microphones 20 the direction of each one of the microphones 20 is identified by a direction of a main receiving lobe of each one of the microphones 20.
[0099] Alternatively it is possible to envisage that the event-based camera 40 is substituted by other means to detect the direction of the speech . By using data fromsaid other means to detect the direction of the speech, the processor identifies a direction from which the speech comes, therefore the processor activates the directional sound amplification step 33.
[0100] More generally is it possible to envisage that said wearable device comprises at least one microphone 20 receiving audio signals, at least one speaker 30, at least one processor and at least one memory comprising hearing aid module .
[0101] Said hearing aid module comprises speech detection functionality, omnidirectional sound amplification functionality, speech direction detection functionality and directional sound amplification functionality .
[0102] The method 100 for operating said wearable device comprises said measuring step 1 provides that said at least one processor measures an intensity of audio signal received by said at least one microphone 20.
[0103] The method 100 comprises said speech detection step 2 provides that said at least one processor detects speech from the sound received by said at least one microphone 20 by using said speech detection functionality .
[0104] The method 100 comprises said omnidirectional sound amplification step 23 provides that said at least one processor modifies amplification levels of said at least one microphone 20 and transmits amplificated audio signal to said at least one speaker 30 by using said omnidirectional sound amplification functionality .
[0105] The method 100 comprises a speech direction detection step 3 provides that said at least one processor detects that the speech direction by using said speech direction detection functionality,The method 100 comprises said directional sound amplification step 33 provides that said at least one processor activates directional sound amplification functionality to amplify the audio signal received within a direction from which the speech comes and transmits the amplificated audio signal from said direction to said at least one speaker 30.
[0106] If measuring step 1 measures that the audio signal intensity is higher than a predefined threshold value, then said at least one processor activates said speech detection step 2.
[0107] If said speech detection step 2 detects the measured audio signal comprises speech, then said at least one processor activates said omnidirectional sound amplification step 23 and said speech direction detection step 3.
[0108] If said speech direction detection step 3 detects that speech direction faces towards the user wearing the wearable device, then said at least one processor activates said directional sound amplification step 33.
[0109] Advantageously, the method 100 for operating said wearable device activates hearing aid module in a context-aware manner, extends the battery life of the wearable device, minimizes power consumption, and eliminates a need for manual activation or deactivation of the hearing aid module by the user .
[0110] More generally said wearable device comprises at least one microphone 20 receiving audio signals, at least one speaker 30, at least one processor and at least one memory comprising hearing aid module .
[0111] Said hearing aid module comprises speech detection functionality, omnidirectional sound amplificationfunctionality, speech direction detection functionality and directional sound amplification functionality .
[0112] Said at least one processor is adapted to measure an intensity of audio signal received by said at least one microphone 20.
[0113] Said at least one processor is adapted to detect speech from the sound received by said at least one microphone 20 by using said speech detection functionality .
[0114] Said at least one processor is adapted to modify amplification levels of said at least one microphone 20 and transmits amplificated audio signal to said at least one speaker 30 by using said omnidirectional sound amplification functionality .
[0115] Said at least one processor is adapted to detect the speech direction by using said speech direction detection functionality .
[0116] Said at least one processor is adapted to activate directional sound amplification functionality to amplify the audio signal received within a direction from which the speech comes and transmits the amplificated audio signal from said direction to said at least one speaker 30 .
[0117] If said at least one processor measures that the audio signal intensity is higher than a predefined threshold value, then said at least one processor is adapted to activate said speech detection functionality .
[0118] If said said at least one processor detects the measured audio signal comprises speech, then said at least one processor is adapted to activate said omnidirectional sound amplification functionality and said speech direction detection functionality .If said at least one processor detects that speech direction faces towards the scene of the user wearing the wearable device, then said at least one processor is adapted to activate said directional sound amplification functionality .
[0119] Advantageously the wearable device activates hearing aid module in a context-aware manner, extends the battery life of the wearable device, minimizes power consumption, and eliminates a need for manual activation or deactivation of the hearing aid module by the user .
[0120] Advantageously the head-mounted device 10 comprises the wearable device that activates hearing aid module in a context-aware manner, that extends the battery life of the wearable device, that minimizes power consumption, and eliminates a need for manual activation or deactivation of the hearing aid module by the user .
[0121] The invention thus conceived is susceptible to many modifications and variants, as long as they fall within the scope of the claims . In practice, the materials used, as well as their dimensions, can be of any type according to the technical requirements .
Claims
- l-CLAIMS1. Method (100) for operating a wearable device, wherein said wearable device comprises at least one microphone (20) receiving audio signals, at least one speaker (30) , at least one processor and at least one memory comprising a hearing aid module,wherein said hearing aid module comprises speech detection functionality, omnidirectional sound amplification functionality, speech direction detection functionality and directional sound amplification functionality,said method (100) comprisinga measuring step (1) provides that said at least one processor measures an intensity of audio signal received by said at least one microphone (20) ,a speech detection step (2) provides that said at least one processor detects speech from the sound received by said at least one microphone (20) by using said speech detection functionality,an omnidirectional sound amplification step (23) provides that said at least one processor modifies amplification levels of said at least one microphone (20) and transmits amplificated audio signal to said at least one speaker (30) by using said omnidirectional sound amplification functionality,a speech direction detection step (3) provides that said at least one processor detects that the speech direction by using said speech direction detection functionality,a directional sound amplification step (33) provides that said at least one processor activates directional sound amplification functionality to amplifythe audio signal received within a direction from which the speech comes and transmits the amplificated audio signal from said direction to said at least one speaker (30) ,if measuring step (1) measures that the audio signal intensity is higher than a predefined threshold value, then said at least one processor activates said speech detection step (2) ,if said speech detection step (2) detects the measured audio signal comprises speech, then said at least one processor activates said omnidirectional sound amplification step (23) and said speech direction detection step (3) ,if said speech direction detection step (3) detects that speech direction faces towards a user wearing the wearable device, then said at least one processor activates said directional sound amplification step (33) .
2. Method (100) according to claim 1, characterized in that said wearable device comprises at least one event-based camera (40) ,wherein said directional sound amplification functionality comprises face detection functionality, face direction identification functionality, lips movement detection functionality,wherein said face detection functionality provides that at least one processor detects a face or at least one region of said face from which the received audio signals come, using data from said at least one eventbased camera (40) ,wherein said face direction identification functionality provides that said at least one processoridentifies the direction of the face from which the audio signal received comes by using data of said at least one event-based camera (40) ,wherein said lips movement detection functionality provides that said at least one processor detects lips movement on said face or said at least one region of said face by using data of said at least one event-based camera ( 0) .
3. Method (100) according to claim 2, characterized in that said speech direction detection step (3) provides thatif said face detection functionality detects a face by using the data of said at least one event-based camera (40) , then said at least one processor activates said face direction identification functionality,if said face direction identification functionality detects that the speech comes from said face, then said at least one processor activates said lips movement detection functionality,if said lips movement detection functionality detects that the lips of the face are moving, then said at least one processor activates said directional sound amplification step (33) .
4. Method (100) according to one or more of anyone of the preceding claims, characterized in that said wearable device comprises at least two microphones (20) spaced apart from each other, and said said directional sound amplification functionality comprises beamforming functionality,wherein said said directional sound amplification step (33) provides that said beamforming functionality provides that said at least one processor applies saidbeamforming to said at least two microphones (20) and constructively combines the audio signal received from the direction where the speech comes .
5. Method (100) according to one or more of anyone of the preceding claims, characterized in that said omnidirectional sound amplification step (23) comprises at least one step comprises in a list comprising modifying amplification levels which provides that said at least one processor modifies amplification levels of said at least one microphone (20) and / or amplification levels of audio signal transmitted to said at least one speaker (30) ,implementing noise reduction algorithms which provides that said at least one processor implements noise reduction algorithm to said at least one speaker (30) .
6. Method (100) according to one or more of anyone of the preceding claims, characterized in that said hearing aid module is implemented with a classifier operated by said at least one processor,wherein said omnidirectional sound amplification step (23) comprises an assessment step which provides that said classifier assesses the sound received and categorizes audio data from sound received into a multiplicity of listening environments, enabling an adjustment of hearing aid module .
7. Method (100) according to one or more of anyone of the preceding claims, characterized in that said directional sound amplification step (33) provides that said at least one processor amplifies and / or combines the sound received, or increases audio intensity of said at least one of said at least one microphones (20) , orincreases at least one receiving lobe of said at least one microphone (20) .
8. Method (100) according to one or more of anyone of the claims 2-7, characterized in that said wearable device comprises a Time of Flight Camera,wherein said speech direction detection step (3) provides that said at least one processor detects said face from which said speech comes by using data from said at least one event-based camera (40) and from a Time of Flight Camera .
9. Wearable device comprises at least one microphone (20) receiving audio signals, at least one speaker (30) , at least one processor and at least one memory comprising hearing aid module,wherein said hearing aid module comprises speech detection functionality, omnidirectional sound amplification functionality, speech direction detection functionality and directional sound amplification functionality,wherein said at least one processor is adapted to measure an intensity of audio signal received by said at least one microphone (20) ,wherein said at least one processor is adapted to detect speech from the sound received by said at least one microphone (20) by using said speech detection functionality,wherein said at least one processor is adapted to modify amplification levels of said at least one microphone (20) and transmits amplificated audio signal to said at least one speaker (30) by using said omnidirectional sound amplification functionality, wherein said at least one processor is adapted todetect the speech direction by using said speech direction detection functionality,wherein said at least one processor is adapted to activate directional sound amplification functionality to amplify the audio signal received within a direction from which the speech comes and transmits the amplificated audio signal from said direction to said at least one speaker (30) ,if said at least one processor measures that the audio signal intensity is higher than a predefined threshold value, then said at least one processor is adapted to activate said speech detection functionality, if said said at least one processor detects the measured audio signal comprises speech, then said at least one processor is adapted to activate said omnidirectional sound amplification functionality and said speech direction detection functionality,if said said at least one processor detects that speech direction faces towards the wearable device, then said at least one processor is adapted to activate said directional sound amplification functionality .
10. Wearable device according to claim 9, characterized in that said wearable device comprises at least one event-based camera (40) ,wherein said directional sound amplification functionality comprises face detection functionality, face direction identification functionality, lips movement detection functionality,wherein said face detection functionality provides that at least one processor is adapted to detect a face or at least one region of said face from which the received audio signals come, using data from said atleast one event-based camera (40) ,wherein said face direction identification functionality provides that said at least one processor is adapted to identify the direction of the face from which the audio signal received comes by using data of said at least one event-based camera (40) ,wherein said lips movement detection functionality provides that said at least one processor is adapted to detect lips movement on said face or said at least one region of said face by using data of said at least one event-based camera (40) .
11. Wearable device according to claim 10, characterized in thatif said face detection functionality detects a face by using the data of said at least one event-based camera (40) , then said at least one processor is adapted to activate said face direction identification functionality,if said face direction identification functionality detects that the speech comes from said face, then said at least one processor is adapted to activate said lips movement detection functionality,if said lips movement detection functionality detects that the lips of the face are moving, then said at least one processor is adapted to activate said directional sound amplification functionality .
12. Wearable device according to one or more of anyone of the claims 9-11, characterized in that said wearable device comprises at least two microphones (20) spaced apart from each other, and said said directional sound amplification functionality comprises beamforming functionality,wherein said said directional sound amplification functionality provides that said beamforming functionality provides that said at least one processor is adapted to apply said beamforming to said at least two microphones (20) and is adapted to constructively combine the audio signal received from the direction where the speech comes .
13. Wearable device according to one or more of anyone of the claims 9-12, characterized in that said omnidirectional sound amplification functionality comprises at least one functionality comprises in a list comprisingmodifying amplification levels functionality which provides that said at least one processor is adapted to modify amplification levels of said at least one microphone (20) and / or amplification levels of audio signal transmitted to said at least one speaker (30) , implementing noise reduction functionality which provides that said at least one processor is adapted to implement noise reduction algorithm to said at least one speaker (30) .
14. Wearable device according to one or more of anyone of the claims 9-13, characterized in that said hearing aid module is adapted to be implemented with a classifier operated by said at least one processor, wherein said omnidirectional sound amplification functionality comprises an assessment functionality which provides that said classifier is adapted to assess the sound received and is adapted to categorize audio data from sound received into a multiplicity of listening environments, is adapted to enable an adjustment of hearing aid module .
15. Wearable device according to one or more of anyone of the claims 9-14, characterized in that said directional sound amplification functionality provides that said at least one processor is adapted to amplify and / or to combine the sound received, or is adapted to increase audio levels of said at least one of said at least one microphones (20) , or is adapted to increase at least one receiving lobe of said at least one microphone (20) .
16. Wearable device according to one or more of anyone of the claims 10-15, characterized in that said wearable device comprises a Time of Flight Camera, wherein said speech direction detection functionality provides that said at least one processor is adapted to detect said face from which said speech comes by using data from said at least one event-based camera (40) and from a Time of Flight Camera .
17. Head-mounted device (10) comprising a wearable device comprising at least one microphone (20) receiving audio signals, at least one speaker (30) , at least one processor and at least one memory comprising hearing aid module,wherein said hearing aid module comprises speech detection functionality, omnidirectional sound amplification functionality, speech direction detection functionality and directional sound amplification functionality,wherein said at least one processor is adapted to measure an intensity of audio signal received by said at least one microphone (20) ,wherein said at least one processor is adapted to detect speech from the sound received by said at leastone microphone (20) by using said speech detection functionality,wherein said at least one processor is adapted to modify amplification levels of said at least one microphone (20) and transmits amplificated audio signal to said at least one speaker (30) by using said omnidirectional sound amplification functionality, wherein said at least one processor is adapted to detect the speech direction by using said speech direction detection functionality,wherein said at least one processor is adapted to activate directional sound amplification functionality to amplify the audio signal received within a direction from which the speech comes and transmits the amplificated audio signal from said direction to said at least one speaker (30) ,if said at least one processor measures that the audio signal intensity is higher than a predefined threshold value, then said at least one processor is adapted to activate said speech detection functionality, if said said at least one processor detects the measured audio signal comprises speech, then said at least one processor is adapted to activate said omnidirectional sound amplification functionality and said speech direction detection functionality,if said said at least one processor detects that speech direction faces towards a user wearing the wearable device, then said at least one processor is adapted to activate said directional sound amplification functionality .
18. Head-mounted device (10) according to claim 17, characterized in that comprises a wearable device according to one or more of anyone of the claims 9-16.
