Sound Output Apparatus and System with Hearing Assistance Function

Abstract

A sound output device and system equipped with a hearing assistance function are disclosed. According to one aspect of the present embodiment, a sound output device having a hearing aid function comprises: a microphone for receiving a noise-voice signal mixed with noise and voice from an external source; an ADC for converting the noise-voice signal converted from the microphone from an analog signal to a digital signal; a hearing aid module for receiving a user's hearing audiogram from an external source and providing a hearing aid function corresponding to the user's hearing audiogram; a DAC for converting a final output signal to which the hearing aid function is applied, provided from the hearing aid module, into an analog signal; and a speaker for outputting the final output signal converted into an analog signal from the DAC as a voice signal.

Description

Technology Field

[0001] This embodiment relates to a sound output device and system equipped with an auditory assistance function. Background Technology

[0002] The content described in this section merely provides background information regarding the present embodiment and does not constitute prior art.

[0003] According to the World Health Organization (WHO), the global population with hearing loss is approximately 466 million, of which 92.7%, or 432 million, are adults. It is projected that the number of people diagnosed with hearing loss will exceed 900 million by the year 6,500. The increase in the hearing-impaired population is leading to a rise in the use of hearing aids, and consequently, the global hearing aid market is recording an average annual growth rate of 6.9%. In Korea, approximately 18,000 hearing aids were sold in 6,107, with an average annual growth rate of 8.5%, exceeding the global average.

[0004] Generally, hearing aids include filter banks that separate or combine incoming sound by frequency band, noise cancellation, acoustic feedback elimination, and wide dynamic range compression, which adjusts the dynamic range of sound appropriately according to the listener's hearing loss characteristics to control amplification.

[0005] However, since hearing aids must be inserted into the wearer's ear, they are designed to be small in size (volume). Consequently, the size of the battery included in a standard hearing aid is also inevitably small, requiring it to operate at low power for extended periods. However, among the functions that a standard hearing aid must perform, the filter bank function, in particular, presents a problem of consuming significant power. As a result, conventional hearing aids have been difficult to use for long periods, causing various inconveniences to users. The problem to be solved

[0006] One embodiment of the present invention has the objective of providing a sound output device and system that improve battery consumption performance and processing speed by having a hearing assistance function. means of solving the problem

[0007] According to one aspect of the present embodiment, a sound output device having a hearing aid function comprises: a microphone for receiving a noise-voice signal mixed with noise and voice from an external source; an ADC for converting the noise-voice signal converted from the microphone from an analog signal to a digital signal; a hearing aid module for receiving a user's hearing audiogram from an external source and providing a hearing aid function corresponding to the user's hearing audiogram; a DAC for converting a final output signal to which the hearing aid function is applied, provided from the hearing aid module, into an analog signal; and a speaker for outputting the final output signal converted into an analog signal from the DAC as a voice signal.

[0008] According to one aspect of the present embodiment, the microphone is characterized by converting a received noise voice signal into an electrical signal.

[0009] According to one aspect of the present embodiment, the hearing aid module is characterized by providing a hearing aid function using a hearing aid learning model.

[0010] According to one aspect of the present embodiment, the microphone is characterized by being implemented in a position that is exposed to the outside when the sound output device is mounted to a user.

[0011] According to one aspect of the present embodiment, the speaker is characterized by being located on the inner side closest to the user's auditory organ within the sound output device when the sound output device is mounted on the user.

[0012] According to one aspect of the present embodiment, the sound output device is characterized by being inserted into the user's auditory organ or located near the user's auditory organ.

[0013] According to one aspect of the present embodiment, the sound output device is characterized by being implemented as an earphone or a headphone.

[0014] According to one aspect of the present embodiment, the hearing aid module comprises a communication unit that receives a hearing audiogram from the outside and a noise speech signal converted into a digital signal from the ADC, a hearing aid unit that generates a final output signal from the noise speech signal converted into a digital signal using a hearing aid learning model trained to take various human hearing audiograms and noise speech signals as input values ​​and output a pure speech signal within the noise speech signal, a speech estimation signal with dynamic range compression applied, and a noise signal as output values, and a memory unit that stores various pure speech signals, various noise signals, and the hearing aid learning model.

[0015] According to one aspect of the present embodiment, the hearing aid unit is characterized by inputting the hearing audiogram and noise voice signal received by the communication unit as input values ​​to the artificial intelligence learning model.

[0016] According to one aspect of the present embodiment, the hearing aid is characterized by combining a speech estimation signal with dynamic range compression applied and an estimated noise signal to output a final output signal. Effects of the invention

[0017] As described above, according to one aspect of the present embodiment, by providing a hearing aid function, not only is the battery consumption performance and processing speed improved compared to conventional hearing aids, but there is also an advantage in that it can be included within a general sound output device rather than a hearing aid and perform the hearing aid function together with the inherent operation of the sound output device. Brief explanation of the drawing

[0018] FIG. 1 is a diagram illustrating the configuration of a sound output system according to one embodiment of the present invention. FIG. 2 is a diagram illustrating the configuration of a sound output device according to one embodiment of the present invention. FIG. 3 is a diagram illustrating the configuration of a hearing assistance module according to one embodiment of the present invention. FIG. 4 is a diagram illustrating the learning process of a hearing aid learning model according to one embodiment of the present invention. Figure 5 is a graph showing the audiogram of a typical listener's hearing. FIG. 6 is a diagram illustrating a dynamic range compression operation performed by a hearing aid according to one embodiment of the present invention. FIG. 7 is a diagram illustrating the configuration and operation of a hearing aid unit according to one embodiment of the present invention. Specific details for implementing the invention

[0019] The present invention is susceptible to various modifications and may have various embodiments, and specific embodiments are illustrated in the drawings and described in detail. However, this is not intended to limit the invention to specific embodiments, and it should be understood that the invention includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the invention. Similar reference numerals have been used for similar components in the description of each drawing.

[0020] Terms such as first, second, A, B, etc., may be used to describe various components, but said components should not be limited by said terms. These terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component. The term "and / or" includes a combination of a plurality of related described items or any of a plurality of related described items.

[0021] When it is stated that one component is "connected" or "connected" to another component, it should be understood that while it may be directly connected or connected to that other component, there may also be other components in between. On the other hand, when it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between.

[0022] The terms used in this application are used merely to describe specific embodiments and are not intended to limit the invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "having" should be understood as not precluding the existence or addition of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification.

[0023] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which this invention pertains.

[0024] Terms such as those defined in commonly used dictionaries should be interpreted as having meanings consistent with their meanings in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application.

[0025] In addition, each component, process, procedure, or method included in each embodiment of the present invention may be shared within a scope that is not technically contradictory to one another.

[0026] FIG. 1 is a diagram illustrating the configuration of a sound output system according to one embodiment of the present invention.

[0027] Referring to FIG. 1, a sound output system (100) according to one embodiment of the present invention includes a sound output device (110) and a terminal (120).

[0028] The sound output system (100) corrects and provides externally input noise audio signals based on the user's hearing audiogram to match the user's hearing audiogram. Here, the noise audio signal is a signal in which noise and voice are mixed, and refers to a normal voice signal provided in a general environment where people live. The sound output system (100) assists the user's hearing by allowing the user to use a sound output device (110) equipped with a hearing assistance function, rather than requiring the user to separately equip a hearing assistance device such as a hearing aid. Since the user can use the sound output device (110), they can use the hearing assistance function without using a hearing aid, which is relatively expensive. Furthermore, because the user can use the sound output device (110) containing a relatively large capacity battery or the sound output device (110) that they have generally been using and frequently charging, they can avoid the additional inconvenience they would feel from using a hearing aid.

[0029] The sound output device (110) performs a hearing assistance function for noise voice signals introduced from the outside, and at the same time outputs a voice signal provided from the terminal (120) into the user's auditory organ. The sound output device (110) is implemented in a form that is inserted into or located near the user's auditory organ, for example, as an earphone or headphones, and provides a voice signal to the user.

[0030] At this time, the sound output device (110) includes a hearing aid function and corrects and provides a noise voice signal entering from the outside to the user to match the user's hearing audiogram. Accordingly, the user of the sound output device (110) may listen to the voice signal provided from the terminal (120) according to their needs, and may also listen to the noise voice signal entering from the outside more smoothly. In particular, the sound output device (110) minimizes the problems of conventional hearing aids by including a hearing aid module (described later with reference to FIG. 2) that performs a hearing aid function but does not operate in a way that consumes significant power like conventional hearing aids. The specific configuration and operation of the sound output device (110) will be described later with reference to FIG. 2 and below.

[0031] The terminal (120) transmits a voice signal (which the user wants to hear of their own volition), the user's hearing audiogram, and a noise level adjustment value to a sound output device (110).

[0032] The terminal (120) communicates with the sound output device (110) via wired or wireless means, receives a voice signal from the outside, or transmits a previously stored voice signal to the sound output device (110) according to the user's input. The terminal (120) receives input from the user regarding the voice signal it intends to output, and transmits the received voice signal to the sound output device (110), thereby causing the sound output device (110) to output the corresponding voice signal.

[0033] The terminal (120) calculates the user's hearing audiogram and transmits it to the sound output device (110). The terminal (120) analyzes the sound intensity perceived by the individual listener for each frequency band and calculates the listener's masking threshold from this. The process of the terminal (110) calculating the masking threshold is exemplified in FIG. 5.

[0034] Figure 5 is a graph showing the audiogram of a typical listener's hearing.

[0035] Referring to FIG. 5, the terminal (120) provides a voice signal to a listener via a sound output device (110), adjusting the volume of the sound for each frequency. The listener who hears the voice signal inputs to the terminal (120) whether they perceive the voice signal they are hearing, and the terminal (120) calculates the listener's masking threshold from the input. Hearing has the characteristic that if the sound intensity (of a specific frequency band) does not exceed the masking threshold, it is not perceived less by the listener in proportion to the difference value, but is consistently not perceived regardless of the difference value. Accordingly, according to the example of FIG. 5(a), the bands of approximately 4200–4300 Hz and 5000 Hz, where the sound intensity is lower than the masking threshold, have the characteristic of not being perceived by the listener.

[0036] Referring again to FIG. 1, the terminal (120) calculates the audiogram of the listener's hearing using a masking threshold and transmits the calculated audiogram to the sound output device (110).

[0037] The terminal (120) receives a noise level adjustment value and transmits it to the sound output device (110). If the user listens only to a pure voice signal that does not contain noise, they may feel a significant sense of unfamiliarity. Accordingly, when a pure voice signal is provided among the noise voice signals introduced from the outside, the noise signal within the noise voice signal introduced from the outside is additionally included and provided. However, if provided as is, there may be cases where the pure voice signal cannot be heard smoothly depending on the state of the user's hearing audiogram. Accordingly, the terminal (120) causes the sound output device (110) to output a signal containing a noise signal in an arbitrary voice signal, and receives a noise level adjustment value from the user regarding how much noise signal is included at a level that is most comfortable for the user to listen to or allows them to perceive the voice signal well without a sense of unfamiliarity. The terminal (120) receives the noise level adjustment value from the user and transmits it to the sound output device (110). The sound output device (110) generates and outputs a final output signal by reflecting the received noise level adjustment value, thereby enabling the user to clearly recognize the pure voice signal among the noise voice signals provided from the outside without feeling any sense of strangeness.

[0038] FIG. 2 is a diagram illustrating the configuration of a sound output device according to one embodiment of the present invention.

[0039] Referring to FIG. 2, a sound output device (110) according to one embodiment of the present invention includes a microphone (210), an ADC (220), a hearing aid module (230), a DAC (240), and a speaker (250).

[0040] The microphone (210) is implemented in a location that is exposed to the outside when the sound output device (110) is mounted on the user, and receives a noise voice signal from the outside. The microphone (210) receives a noise voice signal from the outside and converts it into an electrical signal.

[0041] The ADC (220) converts the noise voice signal converted from the microphone (210) from an analog signal to a digital signal.

[0042] The hearing aid module (230) communicates with the terminal (120) and receives a hearing audiogram from the terminal (120), and provides a hearing aid function that corrects a noise audiogram converted into a digital signal to match the user's audiogram. The hearing aid module (230) includes a configuration to be described later with reference to FIG. 3, thereby providing a final output signal (including a noise signal of the degree (size) desired by the user and a voice estimation signal with dynamic range compression applied) for the noise acoustic signal of the surrounding environment. Accordingly, the hearing aid module (230) performs hearing correction for the user, similar to a conventional hearing aid.

[0043] At this time, the hearing aid module (230) provides a hearing aid function using a hearing aid learning model (artificial intelligence learning model) as described below, thereby significantly improving battery consumption performance and processing speed compared to conventional hearing aids. Accordingly, the hearing aid module (230) is installed within a general sound output device (110), so that the sound output device (110) can perform a hearing aid function while minimizing problems related to battery consumption.

[0044] The DAC (240) converts the final output signal provided from the hearing aid module (230) into an analog signal.

[0045] The speaker (250) outputs the final output signal (with a hearing aid function applied) converted into an analog signal from the DAC (240) as a voice signal. When the sound output device (110) is mounted on the user, the speaker (250) is located on the inner side closest to the user's auditory organ within the sound output device (110) to output the final output signal as a voice signal. By receiving the voice signal (final output signal) output from the speaker (250), the user can clearly perceive external noise voice signals.

[0046] FIG. 3 is a diagram illustrating the configuration of a hearing assistance module according to one embodiment of the present invention.

[0047] Referring to FIG. 3, a hearing aid module (230) according to one embodiment of the present invention includes a communication unit (310), a hearing aid unit (320), and a memory unit (330).

[0048] The communication unit (310) receives a hearing audiogram and a noise level adjustment value from the terminal (120), and receives a noise voice signal converted into a digital signal from the ADC (220).

[0049] The hearing aid unit (320) generates a final output signal from a noise speech signal converted into a digital signal using a hearing aid learning model (artificial intelligence learning model, 450).

[0050] The hearing aid learning model (450) is trained by a separate learning unit to take various human hearing audiograms and noise speech signals as input values, and to output a pure speech signal separated within the noise speech signal, a speech estimation signal with dynamic range compression applied, and a noise signal as output values. The separate learning unit receives various human hearing audiograms and noise speech signals (more specifically, various pure speech signals and pure noise signals) as input values ​​and trains a hearing aid learning model that separates the noise speech signal into a pure speech signal, a noise signal, and a speech estimation signal with dynamic range compression applied according to the hearing audiogram. The training of the hearing aid learning model (450) proceeds as illustrated in FIG. 4.

[0051] FIG. 4 is a diagram illustrating the learning process of a hearing aid learning model according to one embodiment of the present invention.

[0052] Referring to FIG. 4, a separate learning unit receives an arbitrary pure voice signal and an arbitrary noise signal among the signals stored in the memory unit (330). The separate learning unit combines the input pure voice signal and the noise signal to generate an arbitrary noise voice signal that a user typically hears in daily life.

[0053] The hearing aid learning model (450) receives an arbitrary noise speech signal generated as described above as an input value in an architecture such as a DNN model, and learns to output as an output value the result of separating the input arbitrary noise speech signal into a pure speech signal, a speech estimation signal with dynamic range compression applied, and a noise signal, respectively. The hearing aid learning model (450) receives an arbitrary hearing audiogram along with the arbitrary noise speech signal as an input value. The hearing aid learning model (450) separates the speech signal into a speech estimation signal with dynamic range compression based on the pure speech signal, the noise signal, and the hearing audiogram within the noise speech signal.

[0054] Meanwhile, any pure voice signal input to a separate learning unit is input to a filter bank that divides the signal into frequency bands. The filter bank includes multiple (FIR) bandpass filters and divides the applied pure voice signal into different frequency bands. Subsequently, the dynamic compression unit (430) performs dynamic range compression on each frequency divided in the filter bank using the input arbitrary hearing audiogram. Dynamic range compression is performed as illustrated in FIG. 6.

[0055] FIG. 6 is a diagram illustrating a dynamic range compression operation performed by a hearing aid according to one embodiment of the present invention.

[0056] Referring to FIG. 6(a), the range of acoustic power that a general person can hear is the first range (610), and there may be a hearing loss range (620) that a general user has lost. In this case, the dynamic compression unit (430) can perform dynamic range compression by compressing the dynamic range of sound acoustic power from the first range (610) to the second range (630). Accordingly, the dynamic range of sound appearing in the first range (610) can be compressed and expressed in the second range (630).

[0057] However, depending on the case, the user's hearing loss range may manifest in various ways. For example, as illustrated in FIG. 6(b), in the case of other users (e.g., those with autism spectrum disorder), both a sound intensity sensitive range (640) and a sound non-stimulating range (650) may exist. The dynamic compression unit (430) can compress the dynamic range of sound acoustic power from the first range (610) to only the third range (660). Accordingly, the dynamic compression unit (430) compresses the dynamic range of sound appearing in the first range (610) to only the third range (660) so that the user can perceive it.

[0058] Referring again to FIGS. 3 and 4, the dynamic compression unit (430) analyzes the hearing loss area of ​​the user (who possesses the corresponding hearing audiogram) from the hearing audiogram and compresses the voice signals for each frequency into the remaining range excluding the area.

[0059] Subsequently, the filter bank recombines the dynamic range compressed speech estimation signals for each frequency. Through this process, a speech estimation signal with dynamic range compression applied according to the input audiogram is output for the input pure speech signal.

[0060] As described above, the hearing aid learning model (450) takes various human hearing audiograms and noise speech signals as input values, and takes a pure speech signal within the noise speech signal, a speech estimation signal with dynamic range compression applied, and a noise signal as output values. However, separately from this, the hearing aid learning model (450) separately receives a (random) pure speech signal input for the generation of the noise speech signal (input as input value), a speech estimation signal with dynamic range compression applied through a filter bank, and a (random) noise signal as correct values. Subsequently, the hearing aid learning model (450) calculates an error (loss) by comparing the pure speech signal within the noise speech signal, the speech estimation signal with dynamic range compression applied, and the noise signal output as output values ​​with the (random) pure speech signal, the speech estimation signal with dynamic range compression applied through a filter bank, and the (random) noise signal received separately. The hearing aid learning model (450) continuously outputs output values ​​based on input values ​​and continuously performs learning in a direction that reduces the error (between output values ​​and correct values) by comparing it with separately input data (correct values).

[0061] The hearing aid learning model (450) undergoes the aforementioned learning and, when it receives an arbitrary noise speech signal and a hearing audiogram as input, outputs at least a speech estimation signal with dynamic range compression applied and an estimated noise signal from it.

[0062] Referring again to FIG. 3, the hearing aid unit (320) generates a final output signal from a noise speech signal converted into a digital signal using a hearing aid learning model (450) learned by a separate learning unit. The operation of the hearing aid unit (320) is illustrated in FIG. 7.

[0063] FIG. 7 is a diagram illustrating the configuration and operation of a hearing aid unit according to one embodiment of the present invention.

[0064] The hearing aid unit (320) inputs a hearing audiogram received from the terminal (120) and a noise speech signal converted into a digital signal received from the ADC (220) as input values ​​to the hearing aid learning model (450). The hearing aid learning model (450) outputs a speech estimation signal with dynamic range compression applied and an estimated noise signal within the input noise speech signal. However, as previously mentioned, if only the speech estimation signal with dynamic range compression applied is output, the user will feel a sense of unfamiliarity. Accordingly, the hearing aid unit (320) combines the speech estimation signal with dynamic range compression applied and the estimated noise signal and outputs them. When combining the speech estimation signal with dynamic range compression applied and the estimated noise signal, the hearing aid unit (320) may combine them after adjusting the magnitude of the noise signal in proportion to the received noise level adjustment value. The hearing aid unit (320) adjusts the magnitude of the estimated noise signal according to the noise level adjustment value and combines it with the estimated voice signal, thereby enabling the user to perceive the estimated voice signal in the most ideal way. The hearing aid unit (320) generates a final output signal through the aforementioned process.

[0065] Referring again to FIG. 3, the memory unit (330) stores a hearing aid learning model (450) learned by a separate learning unit and can store various pure speech signals and various noise signals to be used for learning.

[0066] The memory unit (330) stores the hearing aid learning model (450) learned by a separate learning unit. In terms of storage, the memory unit (330) can store each data set used by the separate learning unit for learning the hearing aid learning model (450) together.

[0067] In addition, the memory unit (330) stores various pure voice signals and various noise signals so that a separate learning unit can perform learning.

[0068] Since the hearing aid module (230) includes the aforementioned configuration and does not perform the filter bank and motion compression functions in real time every time like a conventional hearing aid, it can provide a hearing aid function with significantly improved battery consumption performance and processing speed compared to a conventional hearing aid. Accordingly, the hearing aid module (230) can smoothly perform the aforementioned operation (hearing aid function) even when installed in a general sound output device (110).

[0069] The above description is merely an illustrative explanation of the technical concept of the present embodiment, and a person skilled in the art to which the present embodiment belongs would be able to make various modifications and variations within the scope of the essential characteristics of the present embodiment. Accordingly, the present embodiments are intended to explain, not limit, the technical concept of the present embodiment, and the scope of the technical concept of the present embodiment is not limited by these embodiments. The scope of protection of the present embodiment shall be interpreted by the claims below, and all technical concepts within an equivalent scope shall be interpreted as being included within the scope of rights of the present embodiment. Explanation of the symbols

[0070] 100: Sound output system 110: Sound output device 120: Terminal 210: Microphone 220: ADC 230: Hearing Assist Module 240: DAC 250: Speaker 310: Communications Department 320: Hearing aid 330: Memory section

Claims

Claim 1 A sound output device having a hearing aid function comprises: a microphone that receives a noise-voice signal mixed with noise and voice from an external source; an ADC that converts the noise-voice signal converted from the microphone from an analog signal to a digital signal; a hearing aid module that receives a user's hearing audiogram from an external source and provides a hearing aid function corresponding to the user's hearing audiogram; and a DAC that converts a final output signal to which the hearing aid function is applied, provided by the hearing aid module, into an analog signal. The hearing aid module includes a speaker that outputs a final output signal converted into an analog signal from the DAC as a voice signal, wherein the hearing aid module comprises: a communication unit that receives a hearing audiogram from the outside and a noise voice signal converted into a digital signal from the ADC; and a hearing aid unit that uses a hearing aid learning model trained to take various human hearing audiograms and noise voice signals as input values, output a voice estimation signal with dynamic range compression applied and an estimated noise signal within the input noise voice signal as output values, and combine and output the voice estimation signal with dynamic range compression applied and the estimated noise signal.A sound output device comprising a memory unit that stores various pure speech signals, various noise signals, and the hearing aid learning model, wherein the noise speech signal input as an input value is generated by receiving and combining the pure speech signal and the noise signal, and the input pure speech signal is input into a filter bank that is separately divided by frequency band, and the hearing aid unit performs dynamic range compression on each frequency divided in the filter bank using the input hearing audiogram, analyzes the hearing loss region of the user possessing the corresponding hearing audiogram from the hearing audiogram, compresses speech signals for each frequency to the remaining range excluding the hearing loss region, and the filter bank recombines the speech estimation signals for each frequency that have been dynamically range compressed to output a speech estimation signal with dynamic range compression applied according to the input hearing audiogram for the input pure speech signal. Claim 2 A sound output device according to claim 1, wherein the microphone converts a received noise voice signal into an electrical signal. Claim 3 A sound output device according to claim 1, wherein the hearing aid module provides a hearing aid function using a hearing aid learning model. Claim 4 A sound output device according to claim 1, wherein the microphone is implemented in a position exposed to the outside when the sound output device is mounted to a user. Claim 5 A sound output device according to claim 1, wherein the speaker is located on the inner side closest to the user's auditory organ within the sound output device when the sound output device is mounted on the user. Claim 6 A sound output device according to claim 1, characterized in that the sound output device is inserted into the user's auditory organ or is located near the user's auditory organ. Claim 7 In claim 6, the sound output device is characterized by being implemented as an earphone or a headphone. Claim 8 delete Claim 9 A sound output device according to claim 1, wherein the hearing aid unit inputs the hearing audiogram and noise voice signal received by the communication unit as input values ​​of the hearing aid learning model. Claim 10 In claim 9, the sound output device is characterized in that the hearing aid unit combines a speech estimation signal with applied dynamic range compression and an estimated noise signal to output a final output signal.

Images