The science behind Spice+ Research, Insights and Studies

Transcription

1 The science behind Spice+ Research, Insights and Studies

2 Insights ContourDesign: Cutting-edge technology calls for cutting-edge design Introduction Innovative new chip technologies for hearing instruments are being introduced to the market at a tremendous rate. Leading manufacturers are regularly rolling out new chip platforms offering incredible possibilities approximately every two years. New chip structures with innovative audiological features are being rapidly developed, and the overall design of the hearing instrument is another element that also requires much attention. Endusers will often rely on their first impressions when choosing a hearing instrument reinforcing the importance of hearing instrument design. An attractive look and a comfortable fit are far more important than some may care to believe. A customer will always select the betterlooking instrument first and there are plenty who will not buy a particular system simply because they do not find it visually appealing. Phonak Spice Generation hearing instruments have combined style and performance to create a complete portfolio of unmatched hearing instruments. This new chip platform will offer some exciting new audiological features and benefits, second generation wireless technology, new accessories and attractive new housings. Cutting-edge technology calls for an equally cutting-edge design: ContourDesign. New technology new design As the leading manufacturer of hearing instruments, Phonak has a continuous commitment to innovation rate. As a result two years after the release of the CORE platform we are now introducing a completely redeveloped chip platform with increased technological performance and enduser benefits. The entire range of our behind-the-ear instruments have been re-designed to further enhance these technological improvements. Accessories have also been extensively remodeled, so elements ranging from remote controls to storage cases, all now conform to the new, coordinated overall image. Comfort is key Various features of the new Phonak BTE hearing instruments have been revised to optimize user comfort (both for improved esthetics and benefits). These revisions include: Smallest compact size possible Improvements in acoustic performance Audiological features that support product design Increased reliability Phonak has made all the housings in each power class noticeably smaller than previous models by making improvements in component development and mountings (Figure 1). Fig. 1: Size comparison between Exélia Art M and Ambra microm housing Eye-catching style It is important for new housings to be stylish and to attract the attention of endusers however the aim was evolution, not revolution. The design of the housings for the new Spice Generation has been changed in a subtle, yet effective, ways. Housings are ergonomically rounded where they make contact with the ear, while all visible elements have a modern, sophisticated and therefore esthetically pleasing look. An example of the benefit of such attention to design is evident in the fact that the CRT housings of our Audéo portfolio have been recognized for their excellence by prestigious expert panels (Figure 2). The innovative Audéo YES housing and the ultra-small Audéo MINI are both Red Dot Design award winners. Audéo SMART and Audéo MINI have also been recognized for superior design with the if Design Award. All new Spice generation instruments incorporate all of these award-winning design features, in addition to new and improved design features.

3 Fig. 2: The award winning Audéo portfolio There is a range of colors available for Spice Generation instruments so that virtually any color preference can be accommodated including 2 and 3 color combinations. An additional metal coating produces a beautiful metallic effect and gives the housing a very sophisticated look (Figure 3). The use of three colors creates the illusion of an even smaller and slimmer hearing instrument. Fig. 3: The real metal coating creates a very attractive and authentic metallic effect The back part of the housing, which is visible behind the ear, is designed to match the individual skin tone or hair color of the hearing instrument user. Its concave effect blends into the natural shadow of the pinna and creates the illusion that the housing is disappearing behind the ear. An additional nanocoating helps prevent build up of dirt by repelling these particles, so they can not accumulate in the housing. Form and function The shape of the new housings was not created solely based on appearance, but also designed with function in mind. Figure 4 presents a comparison between an Exélia Art housing which sits behind the ear and a new housing from the same performance class. The new housing ensures that the front microphone remains in its optimal forward position. Along with a rearrangement of the position of the rear microphone, this design supports excellent directional performance. Fig. 4: Improved positioning behind the ear. It is clear that the upper part of the new housing (below) sits deeper in the crease between the skull and the pinna. It is therefore less visible, but the acoustic function is nevertheless improved. Along with the redesign of the housing, there have also been improvements to the Sound Delivery system dedicated to these instruments. All Ambra BTEs use the same earhooks and SlimTubes and housings in the Ambra and Audéo S series now also have: uniform domes the same microphone shield for all models identical control buttons right/left markings integrated child tamper proof battery lock (optional) As part of our continuous commitment to quality, improvements were also made to other housing components. Microphone covers are an essential component to our hearing instruments. If dirt or moisture enters the microphone inlet, the performance of the hearing instrument deteriorates. Careful attention was given to this element, which not only must be a barrier to dirt and moisture, but must also be robust and easy to clean. New SlimTubes and CRT xreceivers also have improvements to fit and design. Connection points to the housings are not only smaller and therefore easier to change but the connection is robust. The ergonomic shape of both tubes and wires better matches the shape of the ear while the CRT wires are even thinner than previous versions. All these small enhancements add up to substantial improvements in wearing comfort and cosmetics.

4 The new standard domes compatible with SlimTubes and xreceivers have been modified to seat better and more securely. The new closed dome now resembles the open dome and the power dome has been resized. A newly designed wax guard, integrated into all domes (Figure 5) making them easy to clean and requiring less frequent replacement. Fig. 5: Improved wax guard in the standard domes For the first time SlimTubes can now reproduce the desirable horn effect, long employed as a fitter option with standard tubing. This new slim tube option improves the acoustic impedance and results in a better sound transmission especially for the higher frequencies. An impressive 5 db more gain at 6 KHz can be achieved! Safety and reliability While form, function and attractive appearances are all essential considerations, safety and reliability are equally important. Users of hearing instruments range in age from infants through to adults, therefore it is vital to ensure that the battery cannot easily fall out, can be securely and properly inserted and can not be opened by children. These important areas were carefully considered during the redevelopment of the new housings and tamperproof battery doors are now available as an option on all Spice Generation BTE hearing instruments. The reliability of the electronics in a hearing instrument begins with the reliability of its housing. Moisture and debris can enter through the microphone and receiver inlets, and seams between the housing parts are also vulnerable points which could allow moisture into the instrument. Special attention has therefore been paid to producing a very accurately manufactured and well-sealed housing. The nanocoating, mentioned previously, also contributes by repelling moisture, causing it to roll off the housing, thereby preventing potentially critical problems from arising in the first place. Summary ContourDesign is reflected in the launch of the new Ambra series, the flagship of the Phonak Spice Generation. Phonak has developed a completely new housing range for BTEs which sets new industry standards in design and esthetics while also encompassing improvements to detail and reliability (Figure 7). All the new housings are designed in two or three colors, with a wide variety of color combinations available, various detailed improvements have been made to enhance acoustics, adaptability and ease of use all created with increased enduser benefit in mind. The new housings are moistureresistant and nano-coated for improved reliability and durability. Phonak Ambra and Audéo S together present a visually harmonized image which in turn reflects the innovative design and exceptional benefits of the new Spice platform. The whole range of products is supplemented by a new remote control and a redesigned storage case, which paired with the new housings reinforce the commitment Phonak has to technological innovation The outcome of the new ContourDesign approach is extremely small, stylish hearing instruments with enhanced performance and enduser benefits. ContourDesign is perfect for a new generation, the Phonak Spice Generation. Fig. 7: All the new housings at a glance Further innovations Re-designing new housings to convey a coordinated overall image, it naturally led Phonak to extend some of these design concepts to our range of accessories as well. A completely redesigned remote control is available for the Phonak Spice Generation products. In addition a new, attractive storage case has been added which will safely store and transport the new hearing instruments (Figure 6). Fig. 6: New remote control and storage case designs

5 Insights SoundRecover: The importance of wide perceptual bandwidth Summary It is particularly important for people with hearing impairment to be able to perceive and discriminate highfrequency sounds easily and accurately. These signals contain information about speech that benefits intelligibility, especially in some common noisy listening conditions. Clear perception of such sounds can also provide valuable localization cues and specific benefits for speech production. Recently, some advanced digital hearing instruments have been introduced that are claimed to provide extended bandwidth, and therefore improved amplification of high-frequency sounds. However, the bandwidth, measured using electroacoustic techniques, is not necessarily representative of the perceptual bandwidth obtained with real fittings. When the perceptual bandwidth is estimated taking into account the audiogram configuration of each hearing-impaired listener, it can be demonstrated that the greater amplification of high frequencies expected with extended-bandwidth devices is difficult to achieve in practice. In contrast, the Phonak proprietary non-linear frequency-compression scheme, SoundRecover, can effectively extend perceptual bandwidth by improving audibility and discrimination of highfrequency signals. Introduction A critical parameter of any communications system is bandwidth, which characterizes its information-carrying capacity. Access to the Internet, for example, is much faster via a broadband than a dial-up connection. This is mainly because a broadband connection utilizes higher frequencies to convey digital data. In general, bandwidth is defined in terms of the range of frequencies that can be carried by a communication channel. Widening the bandwidth means increasing the frequency range, and thereby enabling more information to be delivered through that channel. However, the audibility of a sound such as a pure tone depends not only on its frequency but also on its level. Consequently, a more useful practical definition of bandwidth would specify the range of frequencies at which tones can easily be made comfortably loud. This is illustrated in Figure 1 (Robinson & Dadson, 1957), which shows the level in db SPL (vertical axis) required to produce the same loudness for tones heard across a wide range of frequencies (solid curve). Figure 1 Equal loudness contours for young (solid curve) and older (dashed curve) listeners with normal hearing. The vertical axis shows a moderate level that is perceived as equally loud across frequency (horizontal axis). In this graph, a tone at 1 khz is shown as having a level of 60 db SPL, which would be comfortably loud for an average listener with normal hearing. To maintain the same loudness as the frequency is changed, the level of the tone would need to be adjusted by less than about 10 db across a frequency range from approximately 80 Hz up to nearly 20 khz. At frequencies below 80 Hz, the level would need to be increased for the same perceived loudness. For example, a tone at 20 Hz would have to be presented at about 100 db SPL to be heard The same concept can be applied to hearing. It is commonly accepted that the normal human auditory bandwidth encompasses the range of frequencies from 20 Hz to 20 khz.

6 as equal in loudness to the 1 khz tone at 60 db SPL. This demonstrates that audible bandwidth depends strongly on the sound level, even for normally hearing listeners. Generally, the effective perceptual bandwidth can be increased by raising the level of sounds. Figure 1 also shows equal-loudness data for older listeners who were assumed to have normal hearing (dashed curve). Although those listeners had no signs of ear disease, their average sensitivity to high-frequency tones was much poorer than that of younger listeners (solid curve). At 10 khz, for instance, the level difference between these two groups was almost 20 db for the same loudness. Even larger differences are evident at higher frequencies. In contrast, the listeners age had no effect on the equal-loudness data for frequencies lower than about 2 khz. These measurements are consistent with the findings of many research studies which have shown that high-frequency hearing sensitivity tends to decline as a person ages, even in the absence of any specific pathology. Thus, when the bandwidth of hearing becomes narrower as a result of age-related hearing loss, the usual reason is a change in sensitivity at high frequencies rather than low frequencies. Furthermore, a similar type of bandwidth reduction can result from many common causes of hearing impairment, including exposure to excessive amounts of noise, various diseases, side-effects of ototoxic drugs, and other etiologies. How do these considerations apply to a person who uses a hearing instrument (HI)? The answer is complicated by the presence of two interacting factors. First, there is the particular configuration of each HI user s hearing impairment, as characterized by the audiogram. The second factor is the effective bandwidth of the HI, which depends on its gain and maximum output level, parameters that inevitably vary as a function of frequency. In addition, certain sound-processing techniques such as frequency lowering can affect the perceptual bandwidth. As discussed below, to realistically determine the bandwidth of sounds available to a given HI user it is essential to consider the combined effect of these factors. Perceptual importance of high frequencies Many sounds that contribute to speech intelligibility contain or are dominated by high-frequency components. As just one familiar example, the presence or absence of the phoneme /s/ at the end of almost any English noun indicates whether the speaker means several items or only one item. Depending on the age and gender of the speaker, that phoneme typically has a spectral peak between 4 6 khz, and often contains intense components up to beyond 10 khz. There are numerous other speech sounds in every language that can be discriminated more readily when high-frequency parts of the signal are clearly audible. When a listener is attempting to understand speech in a noisy environment, these acoustic signals are particularly important because they are less susceptible to masking by the relatively intense low-frequency components of many common types of noise. Furthermore, young children with a hearing impairment who are learning a language for the first time benefit from being able to hear the highfrequency speech sounds that they are trying to produce (Stelmachowicz et al, 2002). In addition to these well-established benefits for speech perception (Simpson et al, 2005) and production, ensuring the audibility of high-frequency sounds provides other advantages. For example, some valuable information about the source of sounds, such as birdsong and various important environmental noises, is conveyed principally by high-frequency components. The subjective quality of these sounds tends to be judged as relatively poor if the high frequencies are too soft or inaudible (Moore & Tan, 2003). The ability of people with a hearing impairment to localize sounds that contain high frequencies may also be improved with extended HI bandwidth, because the difference in level of sounds between ears can provide a strong cue to the direction of a sound source. As the level difference must be perceived as a loudness difference between ears for this cue to be reliable, the HI requires adequate bandwidth to ensure high-frequency signals are heard at appropriate levels (Dubno et al, 2002). Hearing-instrument bandwidth In the past, the high-frequency bandwidth limit of analog hearing aids usually resulted mainly from the electroacoustic performance. With high-powered aids in particular, it was often difficult to obtain adequate sound output levels at frequencies above about 4 khz. In recent years, however, receiver technology has improved to the extent that bandwidth limitations are imposed instead by other factors. In all digital hearing instruments, there is an absolute limit on bandwidth resulting directly from the sampling process. Sampling is required to convert the sound signals at the input of the HI into a stream of separate digital representations. The sampling rate has to be high enough to ensure that the continuously varying acoustic signal is represented in the digital processor with adequate fidelity. The selection of sampling rate is based on a fundamental principle of digital signal processing which states that the highest frequency that can be represented adequately after sampling is slightly less than half the sampling rate. For normal hearing listeners, the upper frequency limit is generally assumed to be 20 khz, so the required sampling rate is more than 40 khz. In fact, digital sound recorded using the standard compact disc (CD) format is sampled at a rate of 44.1 khz. Unfortunately, the use of relatively high sampling rates can have undesirable side-effects. The digital signal processor inside any modern hearing instrument is programmed to modify the sound signals at a rate that is equal or

7 proportional to the sampling rate. One practical effect of this relationship is that higher sampling rates cause higher power consumption, and therefore poorer battery lifetime. Designers of digital HIs are faced with a difficult trade-off: widening the acoustic bandwidth of the device means shortening the battery lifetime. Consequently, it is common for the sampling rate in hearing instruments to be approximately 20 khz. This choice means that the upper limit of the bandwidth in terms of sound produced by the HI must be about 10 khz. In some devices, the sampling rate may be as low as 16 khz resulting in an acoustic bandwidth of less than 8 khz. There are several conventional methods of measuring the bandwidth of hearing aids. One widely used technique is specified by the American National Standards Institute (ANSI). In ANSI S3.22, the HI is adjusted to provide amplification in a predetermined reference condition (reference test gain), and the resulting response is measured as a function of frequency. Figure 2 shows a typical measurement, using two hearing instruments with broad bandwidth as an example. For an input at 60 db, the output is averaged at three specific frequencies (usually 1.0, 1.6, and 2.5 khz). Subsequently two frequencies are identified at which the output is 20 db below the calculated average. Those two frequencies are taken to define the lower and upper limits of the bandwidth. For the response curve shown in Figure 2, the bandwidth of Instrument A, estimated according to the ANSI method, is from below 100 Hz to approximately 7.5 khz. demonstrate that bandwidth measurements conducted in accordance with a technical standard do not necessarily provide useful information about the effective bandwidth of a HI when it is fitted to a user. In contrast, a determination of perceptual bandwidth, taking into account not only the electroacoustic characteristics of the HI but also the user s configuration and degree of hearing impairment, is much more informative. Perceptual bandwidth A conventional audiogram records a person s threshold of hearing at a number of discrete frequencies. The lowest frequency is usually 125 or 250 Hz, while the highest frequency may be up to 8 khz. For several technical and practical reasons, it can be difficult to obtain reliable thresholds for very high frequencies (e.g., above 8 khz). Even when threshold levels are available beyond the typical frequencies measured in routine clinical practice, prescriptive fitting rules that specify suitable gain and amplitude compression characteristics for a HI generally do not provide targets at those frequencies. Nevertheless, it would be necessary to know the high-frequency thresholds in order to assess the full range of frequencies that a particular HI is able to make audible when fitted to each individual. Figure 2 Example of ANSI standard bandwidth calculation for two current hearing instruments. Each curve shows the output versus frequency for the reference test gain condition with an input of 60 db. The response is averaged at three frequencies (vertical yellow lines), and then reduced by 20 db (horizontal dashed lines). The bandwidth is delimited by the two frequencies at which these lines intersect the curve. Thus, Instrument A has an upper bandwidth limit of approximately 7.5 khz, whereas Instrument B has an upper limit of 9.2 khz. Figure 2 additionally shows the same measurement for Instrument B. In this case, the bandwidth, determined using the ANSI method, has an upper limit of approximately 9.2 khz. However, it is also clear that the calculated average output of Instrument A is higher than that of Instrument B at every frequency. In fact, if absolute output level rather than the output relative to the reference condition is used to estimate the bandwidth, these two HIs have upper frequency limits that are almost identical. These observations Figure 3 Audiogram for a typical sloping mild to moderate hearing loss Figure 3 is an example of a typical sloping hearing loss of mild to moderate severity, with thresholds at and above 4 khz of 50 db HL. After conversion to equivalent levels at the eardrum, this audiogram is shown as the red curve in Figure 4. Also shown in the latter figure is a fitting of a Phonak HI with wide bandwidth. The proprietary frequency-shifiting algorithm SoundRecover is disabled (green curve). The HI was adjusted to approximate as closely as possible the target recommended by the DSL Adult v5.0a formula.

8 Figure 5 How SoundRecover can extend the perceptual bandwidth. The upper bar shows the full frequency spectrum of sounds at the input of a hearing instrument. Signals with frequencies above the bandwidth limit, shown to the right of the solid vertical line, are not audible to the HI user. With SoundRecover enabled, however, signals above the cut-off frequency (vertical dashed line) are compressed in frequency so that they fall within the available bandwidth (lower bar). Figure 4 The results of fitting two HIs according to the DSL v5.0a formula (green crosses) for the audiogram (red curve) shown in Figure 3. The Phonak HI (green curve) had SoundRecover disabled. The yellow curve shows comparable results from a different manufacturer s HI which is claimed to provide extended bandwidth. It is evident that the Phonak HI without SoundRecover was able to provide useful audibility of the test signal (speech at an average level of 65 db SPL) up to at least 6 khz. The yellow curve in the same figure shows, for comparison, results for a premium-level competitive product which claims extended bandwidth to 10 khz. The measurements plotted in Figure 4 demonstrate clearly that these two HIs result in almost identical perceptual bandwidths when fitted to suit a common audiogram configuration. However, neither HI would provide useful audibility for frequencies higher than about 6 khz, in spite of the fact that the maximum available gain for those frequencies was selected in each device. It is noteworthy that this restriction on audibility above 6 khz is present even for a mild to moderate hearing loss with thresholds in this region of only 50 db HL. This limitation on perceptual bandwidth is a consequence of particular characteristics of both the audiogram and the technical performance of the HIs when fitted for that audiogram. What can be done to overcome this limitation? Currently, the only practical solution is the use of a sophisticated frequency shifting algorithm, which can improve the audibility of highfrequency sound signals without affecting signals at lower frequencies. Unique to Phonak, SoundRecover expands the perceptual bandwidth available to HI users by compressing and shifting a selected input band restricted to high frequencies. The effect of SoundRecover on bandwidth is illustrated in Figure 5, which shows how the maximum input frequency is reduced to fall within the useful bandwidth of the HI when it is fitted appropriately to a person with hearing impairment. Only frequencies above a specified cut-off frequency are compressed in this way. As lower-frequency signals do not pass through the frequency-compression processing, the quality of sounds delivered to the HI user is preserved. A number of research studies have confirmed that speech intelligibility is often improved, both in quiet and in noise, with use of SoundRecover, and that the sound quality of the processing is readily accepted (Glista et al, 2009, Wolfe et al, 2009). These benefits have not been found to be limited to any specific age group, degree of hearing loss or range of audiometric configurations. Figure 6 shows the expected perceptual effects of SoundRecover when enabled in the Phonak hearing instrument. In contrast to Figure 4, this figure shows the output of each HI for a test signal consisting of a noise band centred on 6.3 khz. (This is a synthetic signal, recently made available for clinical use in the Verifit verification system, with characteristics similar to that of the phoneme /s/). Figure 6 As for Figure 4, but for an input signal consisting of a narrow-band noise centred on 6.3 khz. The blue curve shows the effect of enabling SoundRecover in the Phonak HI. Without SoundRecover (green curve), only marginal audibility for this signal can be achieved, and the competitive device (yellow curve) peaks below the hearing threshold and thus does not provide audibility at all. Note that the fitting parameters of each HI remained as described for figure 5, fine-tuned with maximum high frequency gain. With SoundRecover enabled, the test signal is amplified by the Phonak HI to clearly audible levels (blue curve). For further

9 details on how to conduct and interpret this Verifit procedure designed to verify the performance of hearing instruments with frequency shifting technology, please refer to the document Guidelines for fitting hearing instruments with SoundRecover available at In summary, advances in signal processing and receiver design have made it possible to design hearing instruments with an electroacoustic frequency response out to about 10 khz, when measured in a coupler. However, usable real world gain above 6 khz is often not practically achievable even when such devices are fit for a mild to moderate hearing loss. In many actual fittings, the wide bandwidth of the HI itself is not sufficient to extend the perceptual bandwidth and thereby make high-frequency signals audible. Research has shown that the perception of these signals is very important. SoundRecover can provide otherwise unachievable high frequency audibility by extending the perceptual bandwidth of Phonak HIs in addition to the comparatively wide bandwidth already provided by their fundamental electroacoustic design. The benefits of this technology have been scientifically proven by a series of studies published in both peer reviewed and non-peer reviewed journal (see additional publications on SoundRecover at the end of this document). References Dubno JR, Ahistrom JB, Horwitz AR (2002) Spectral contributions to the benefit from spatial separation of speech and noise. J Speech Lang Hear Res 45: Glista D, Scollie S, Bagatto M, Seewald R, and Johnson A (2009) Evaluation of nonlinear frequency compression: Clinical outcomes. Int J Audiol, 48(9): Moore BC, Tan CT, (2003) Perceived naturalness of spectrally distorted speech and music. J Acous Soc Am 114: Robinson DW, Dadson RS (1957) Threshold of hearing and equal-loudness relations for pure tones, and the loudness function. J Acous Soc Am 29: Simpson A, McDermott HJ, Dowell RC (2005) Benefits of audibility for listeners with severe high-frequency hearing loss. Hear Res 210: Stelmachowicz PG, Pittman AL, Hoover BM, Lewis DE (2002) Aided perception of /s/ and /z/ by hearing-impaired children. Ear Hear 23: Wolfe J, Caraway T, John A, Schafer E, & Nyffeler M (2009) Initial experiences with nonlinear frequency compression for children with mild to moderately severe hearing loss. Hear J 62(9): Written by: Prof. Hugh McDermott Deputy Director (Research) The Bionic Ear Institute Albert Street East Melbourne VIC 3002 Australia Additional publications on SoundRecover Bagatto M, Scollie S, Glista D, Pasa V, Seewald R (2008) Case study outcomes of hearing impaired listeners using nonlinear frequency compression technology. Audiology Online, March. Boretzki M, Kegel A (2009) The benefits of nonlinear frequency compression for people with mild hearing loss. Audiology Online, November. Dewald N (2009) Experiences with a wide application of SoundRecover, non-linear frequency compression. Audiology Online, October. Glista D, Scollie S, Polonenko M, Sulkers J (2009) A comparison of performance in children with non-linear frequency compression systems. Hearing Review, November, Kegel A, Boretzki M (2009) Nutzen von SoundRecover für Menschen mit einer milden Hörminderung. Hörakustik August. McDermott HJ, Glista D (2007) SoundRecover: A breakthrough in enhancing intelligibility. Background Story, Phonak AG. McDermott HJ (2010). The benefits of nonlinear frequency compression for a wide range of hearing losses. Audiology Online, January. Nyffeler M (2008) The Naída power hearing instrument family field test results demonstrate better speech clarity unparalleled in its class. Audiology Online, September. Nyffeler M (2008) Study finds that non-linear frequency compression boosts speech intelligibility. Hear J 61(12): Simpson A, Hersbach AA, McDermott HJ (2005) Improvements in speech perception with an experimental nonlinear frequency compression hearing device. Int J Audiol 44(5): Simpson A, Hersbach AA, McDermott HJ (2006) Frequencycompression outcomes in listeners with steeply sloping audiograms. Int J Audiol 45(11): Wolfe J, Caraway T, John A, Schafer E, & Nyffeler M (2009) Verbesserung beim Erkennen und Erlernen hochfrequenter Signale. Hörakustik Oktober.

10 Insights Spice+ Signal Processing: Hearing delight from the start Once digital hearing systems became established at the end of the 1990s, the open-platform principle was introduced as an innovative concept, in contrast to the hard-wired systems (Latzel, 2001). Unlike the hard-wired systems, this concept makes use of freely programmable DSPs (digital signal processors) that offer more flexibility for hearing care professionals and manufacturers. Reprogramming the DSP made it possible not only to adapt the individual parameters of an algorithm to the individual requirements of a person hard of hearing, but also to rework the entire algorithm. Earlier, the algorithms increased in line with the demands for new hardware, which meant that new algorithms also required new signal processors on which to operate them. This meant that the benefits of the open-platform system could never be fully brought into play. For the first time, Spice+ Processing makes use of the openplatform principle in its true sense. By simply reprogramming the DSP and thus updating the hearing system, a new hearing system is created, based on algorithms that differ significantly from those of the original Spice generation. These changes involve far more than mere adaptation of the algorithm parameter, as the core of some algorithms has been revised. The technical and audiological background to Spice+ is outlined in detail below. The Phonak processing philosophy has always involved natural response and good sound quality, with the best possible speech clarity. In developing Spice, the engineers mainly focused on the performance of the hearing systems in acoustically difficult environments such as the classic cocktail party situation. This resulted in special features such as StereoZoom, auto ZoomControl and UltraZoom, which have proven to offer very good speech comprehensibility and impressive sound quality in a noisy environment. The aim of the Spice+ research was to achieve a similarly high level, even in calm acoustic situations. The following scenarios, for example, would be placed in the "calm situations" category: A quiet conversation without any background noise, a quiet conversation in a slightly noisy environment (a hissing fan noise, such as from an air-conditioning unit, etc.), as might be experienced during a business meeting or a lecture, reading the newspaper with or without music from the radio or a ticking clock in the background in the living room, walking in the forest with the leaves gently fluttering. It is easy to see that the "calm situations" category is far more heterogeneous than it might first appear. The idea behind the innovative algorithm that forms part of Spice+ Processing is to automatically provide the optimal setting and the best performance for the various "calm situations". Expansion The response and thus the output of a hearing system are mainly characterised by their electro-acoustic components, the filter bank and the input/output gain curve. If the microphone, receiver and filter bank are regarded as given, it is the dynamic behaviour of the amplifier that mainly determines the perceived sound quality of the hearing aid. The dynamic behaviour of the amplifier is determined by the response curve (input/output curve) and the time constants of the system. As is widely known, the response curve can be divided into 4 ranges (Figure 1): Expansion, linear section, compression (AGC) and limitation.

11 Figure 1: Response curve (output against input) for a typical amplifier with four ranges, integrated into a hearing system: Expansion, linear, compression, limitation (Source Hörakustik - Theorie und Praxis: Hoffmann and Ulrich, 2007) It is mainly the expansion range that is relevant for processing calm situations; this also plays a special role in Spice+ signal processing. Expansion is the opposite of compression, i.e. gain is reduced as the input drops. Expansion is also known as (soft) squelch and is, for example, used in radio technology to suppress constant background noise. Expansion is also very effectively used in many hearing aids to reduce the audibility of input signals at a very low level (Dillon, 2001). This generally relates to the internal noise of the hearing system, primarily caused by the installed electret microphone. Three parameters can be used to adjust expansion: threshold kneepoint (TK), slope and time constants. The correct parameterisation of expansion represents a major challenge to the manufacturer, in endeavouring to ensure that only undesirable soft input signals are filtered out. Manufacturers, therefore, generally do not allow hearing care professionals to have access to the expansion parameters. The Phonak Target 2.0 fitting software, on the other hand, provides hearing care professionals with access to the expansion control threshold by changing the gain level when the input is low, as is explained in detail below. This is a popular means to have a direct effect upon the audibility of soft input signals. When using objective measuring methods, especially when the results are evaluated by means of percentile analysis, the influence of the expansion control level can be very easily detected, as the 30th percentile is directly related to the expansion threshold. Figure 2: Comparison of response (gain against input) for the two hearing aid generations, Spice and Spice+. The input of internal noise is treated in the same way in both cases. However, the different compression ratios result in great differences as regards the TK control level, which is clearly higher for Spice+. With Spice+ Processing, the response has been newly defined in the expansion range. Figure 2 shows the response (shown as gain against input) of Spice+ when compared with Spice. One can see that the curve is clearly flatter (a slope of 0.9 for Spice+, compared with 1.5 for Spice). This results in a clear reduction of the background and internal noise and the device becomes significantly quieter in calm environments with modulated noise (such as a ticking clock), without so-called noise floor. The time constants are another decisive factor when it comes to the perception of modulated noise at a level close to the expansion threshold (T K ). Consequently, the expansion time constants for Spice+ have been fundamentally changed: as a result, both the time constants for rising and falling input thresholds are very rapid, so that not only do soft speech components quickly become audible, but also soft noise components are rapidly reduced. The new time constants significantly affect sound perception, so that hearing aids with Spice+ sound clearly clearerthan Spice systems. Direct Sound Compensation The Direct Sound Compensation algorithm is used for sound quality enhancement in the case of large vents and/or an entirely open fitting. In these cases, the direct sound components that pass through the vent into the ear canal are superimposed on to the amplified sound components (processed by the hearing system). If the phases are unfavourable, these results in comb filter effects, which are experienced by the hearing system user as an echoing sound. In order to counteract this effect, DSC (direct sound compensation) was already introduced in the Spice Generation. This algorithm estimates the proportion of direct and amplified sound in real time. If the calculation gives such an unfavourable ratio that it would result in unnatural sound, the gain of the hearing system is reduced, depending on the frequency and level.

12 In Spice+, the time constants for DSC have been adapted in such a way that the system does not respond too quickly, which would make it appear unsettled, but still kicks in quickly enough to ensure that the sound remains natural in all cases. Figure 3: Example of hearing loss (mild to moderate) Presetting Phonak Target 2.0 offers the option to optimally adapt the system used in Spice+. Three main components have been further developed for this purpose: Presetting of the expansion control threshold (TK controller) Lowering the compression ratio results in a rise in the control threshold. Figure 4 shows a comparison of the control thresholds via the frequency for both Spice and Spice+. Within the 750 Hz to approx. 4 khz frequency range, the control levels for Spice+ Processing are clearly above those of Spice. This measure considerably reduces the gain for low input levels, which has a positive effect upon the noise behaviour of the Spice+ hearing system. Figure 5 shows the output level of Spice and Spice+ when there is no input signal. The output signal is thus only determined by the internal noise of the hearing system. The figure shows that the changes in the control threshold mainly take effect in the case of medium frequencies, whilst the control threshold T K for Spice+ is higher. Precalculation of the individual hearing loss shown (Figure 3) thus reduces the output signal, i.e. the noise, by up to 20 db for Spice+ signal processing. T K s can be set differently for the various programmes and are automatically adapted in SoundFlow. Figure 4: Diagram of the level values as a function of the frequency to which the TK control threshold has been preset in the event of hearing loss according to Figure 3. Figure 6a: Response curve for Spice+. An expansion threshold value TK1 results when low input levels at a specified value are amplified at a given rate (Gain20). Figure 5: Output signal of a hearing system with Spice and Spice+, preset to the hearing loss as shown in Figure 3, without an input signal. The output signal is only determined by the internal noise of the hearing system. The output level of Spice+ is significantly lower at mid frequencies than with Spice Figure 6b: Response curve for Spice+. The value of Gain20 is increased in comparison with the presentation in Figure 6a. As the incline of the expansion curve did not change, increasing this gain results in a reduction of the expansion control level to the value T K2. TK and Gain20 are thus inversely proportional to one another.

13 Note: In Target 2.0, the T K s can only be indirectly adapted individually to the requirements of the hearing system user by varying the gain level for low input levels (Gain20, i.e. gain at an input level of 20 db). Figure 6a shows the response curve (gain of the input level) for a given Gain20. This results in T K1. If the Gain20 is now increased (Figure 6b), a new T K2, which is lower than T K1, results from a constant incline of the curve. This shows that T K and Gain20 are inversely proportional to one another. Adapting the vent loss compensation An open fitting means that gain can be poorly applied at low frequencies, thus - depending on the diameter of the vent - additional gain is required to compensate for the outflow. Depending on the choice of acoustic parameters, i.e. the receiver, vent size and tube length, in Target 1.2, the gain was corrected in such a way that the loss caused by the vent is compensated. For Spice+, Target 2.0 now determines the vent loss compensation not only on the basis of the acoustic parameters, but also takes into account the hearing loss. Where the hearing loss is mild, no gain is required and the loss is also not compensated by the vent. However, where the hearing loss is such that gains are also required for low frequencies, the vent loss is compensated up to 20 db. Figure 7 shows a comparison of the vent loss compensation for Spice and Spice+, given the hearing loss shown in Figure 4 and an open fitting. The measuring difference for this hearing loss is up to 17 db, with the values at the ear obviously not differing that much due to the superimposition of direct sound and amplified sound. Summary The open-platform principle of the Phonak Spice generation makes it possible to create a new hearing aid generation (Spice+) by means of a simple firmware update. As a result, it has been possible to extend the development focus from understanding speech in particularly difficult environments to ensuring good performance in calm situations. Spice+ Processing operates with a new expansion algorithm that softens the sound of the hearing aid and clearly calms its dynamic behaviour. The new first fit calculation, which shifts the control level for the new expansion (T K ) to higher input levels whilst also adapting the vent loss compensation to the hearing loss, guarantees a very high level of initial acceptance for Phonak's new generation of hearing aids. Literature reference Dillon, H. (2001) Hearing Aids, Boomerang Press. Latzel, M. (2001) Kommunikationsprobleme von Hörgeräteträgern bei der Telefonkommunikation - Ansätze zu deren Objektivierung und Lösung, Dissertation Universität Gießen. Ulrich J. and Hoffmann, E. (2007): Hörakustik - Theorie und Praxis, DOZ-Verlag. This article was written by Matthias Latzel, PhD, Head of Audiology at Phonak Germany. For more detailed information, please contact audiology@phonak.com Figure 7: Diagram showing the effect of the vent loss compensation for Spice and Spice+ signal processing when entering the hearing loss given in Figure 3. MPO pre-settings Phonak Target 1.x made direct use of discomfort level input, once it has been converted from a narrow-band to a broadband value, in order to set the MPO. This has often resulted in MPO settings with very high input levels and - especially in the high frequency range - frequently results in MPO settings that are at the limit of the hearing aid. In Phonak Target 2.0, MPO and compression are scaled in a ratio between the measured and the estimated discomfort threshold, so that the remaining dynamic range is optimally used.

14 Insights Phonak Spice Generation Processing: A new generation of sound classification and directionality Introduction Is bigger better or is smaller superior? There is no single correct answer of course as it depends on the dimension to which you are referring. One dimension where bigger is clearly better is the processing power of a Phonak platform - the technology foundation underpinning a generation of hearing instruments spanning chip set, signal processing features, mechanical design and fitting software. The new Phonak Spice Generation platform delivers twice the processing power of the previous generation CORE platform. This extra power has already been tapped by the first wave of Spice hearing instruments and provides a launching pad for advanced sound processing and innovative hearing instruments. Is there such a thing as too much processing power? Have you ever heard anyone say, thanks but my computer processor is actually fast enough, please keep those additional Gigahertz for yourself, oh and I also have plenty of memory as well so you can leave that spare memory stick in the cupboard. While you re at it, you can return that extra 100 gigabyte hard drive to the shop. These comments are never spoken. When the rest of the Spice development team started formulating ways to enhance the performance for wearers and fitters alike, the Phonak hardware engineers quickly responded with the blueprint for a new platform with twice the processing power of CORE. From these solid foundations and sharing a common vision, ongoing collaboration by the entire Phonak global research and development team led to the groundbreaking new Spice platform. Often the best way to appreciate something new is to compare it with an existing point of reference. Figure 1 shows the evolution of key physical dimensions over the last three Phonak platforms. The number of transistors trends upward with each platform generation and the structure trends down. The more transistors available, the more processing can be done. The thinner the structure, the more components can be fit on the microchip and the less power the microchip consumes for the same processing task. Figure 1 - Evolution of key physical dimensions Figure 2 shows a similar picture for some key processing dimensions. The almost doubling of MOPS (million operations per second) from CORE to Spice means the new generation chip does twice as much. Figure 2 - Evolution of key processing dimensions Phonak hearing instruments truly lead the way in taking full advantage of the underlying platform components on which they are built. This Phonak Insight explores some of the signal processing innovations now available on the Spice platform thanks to this leap in processing power.

15 Pinpoint sound environment classification Consider for a moment opening night of a new performance at the city concert hall and specifically the perspective of three different people in attendance. Firstly the usher showing people to their seats. He is busiest the ten minutes before curtain call, but there are always latecomers who need to be seated once the show has started. It is important for the usher to be able to understand a latecomer clearly so as to minimize any disruption for the already seated audience members. Now consider a second person, a music lover with season tickets for the city concert hall. She takes great delight in attending opening night and savoring the first polished performance following weeks of rehearsals. Finally consider a third person in attendance, the gaffer at the concert hall who must ensure the spotlight is always shining on the lead vocalist. He is situated directly above the loudspeakers and must contend with the booming sound whilst maintaining the precise direction of the spotlight. What may initially appear to be the same sound environment is very different for each person, and furthermore each person has a different listening intention. Spice processing provides sophisticated sound classification and intuitive learning capabilities to accurately match a wearer s listening intention to their immediate sound environment. Is accuracy the same as precision? Phonak Spice Generation hearing instruments enjoy the privilege of previously unimaginable processing capability. One of the many features of Spice sound processing that leverages this capability is the sound environment classification of SoundFlow, the automatic system available with every Phonak hearing instrument (Nyffeler, 2009) 1. Most modern hearing instruments do a good job at classifying basic sound environments accurately. However the level of detail, or precision, of this classification is a finer art. SoundFlow classification by the Spice chipset is accomplished by calculating 46 different parameters of the incoming sound such as signal-to-noise ratios, low frequency levels, onsets and spectral roll-offs. These parameters are then analyzed, further combined and projected onto a point within a 3D sound model containing four sound environment spheres - Calm, Speech in Noise, Noise and Music as shown in Figure 3. If a point lies within one of the sound environment spheres it is considered to be 100% associated to that sound environment and the unblended base program for that environment is used. If the point does not lie within any of the spheres, the sound environment association is calculated using a law of attraction. This method provides a natural mapping of a sound environment onto the 3D sound model providing a more accurate, reliable and faster classification of incoming sounds. Figure 3 - High-definition, multi-dimensional sound classification system used by SoundFlow for precise, seamless automatic operation If these calculations don t sound complex enough, the SoundFlow environment classifier now takes into account spatial sound differences. This means that the classifier takes into account which way the wearer is facing. So in the concert hall, speech from inconsiderate audience members sitting behind the music lover is rightly treated as background noise and does not result in any transition to Speech in Noise classification. This precision classification is continuously recalculated in real time enabling SoundFlow to assemble an optimal blended program drawing from the multiple base programs for that instant. Furthermore the transitions between these blended programs are so smooth they are not even noticeable to the wearer. A coming of age in hearing instrument understanding Precise sound environment classification in real time appears very impressive, but to realize a tangible benefit for the wearer their listening intention must also be considered. Spice processing represents a coming of age in the level of understanding a hearing instrument develops with its wearer. FlexControl, a Spice processing breakthrough innovation in intelligent user interaction, allows adjustment along the dimensions of greater clarity or increased comfort as per the wearer s listening intention. Back at the city concert hall - the usher s listening intention is to understand speech in a significant amount of background noise, the music lover s intention is the most pleasurable music experience possible while the gaffer is seeking comfort in what could be an overwhelming noisy situation to allow him to concentrate

16 on the job at hand. Simple volume up/down adjustments are only so useful for the wearer to express their hearing intention. FlexControl performs intelligent adjustments to both gain (hearing loss configuration, desired frequency response and volume levels) and sound cleaning (directional settings, wind noise management, background noise reduction and echo cancellation). Validation results show that FlexControl comprehensively performs better in all sound conditions compared to a conventional volume control (Phonak AG, 2010) 2. User Preference Learning brings together the precision sound classification of SoundFlow and the multi dimensional parameter adjustments made through FlexControl. This convergence of interactive control and automatic adaptation provides an unparalleled new level of hearing instrument control for the wearer. Phonak Spice Generation processing avoids such pitfalls that can affect other systems by separating input level detection from the calculation of spectral amplification. By calculating these parameters many hundreds of times per second independently in both paths, the spectral cues are more accurately reproduced to keep pace with the ever-changing sound environment. So the appropriate level of compression is instantly applied resulting in no gain overshoot as shown by the green line in Figure 4. On opening night, this means the gaffer comfortably experiences the booming opening notes of the performance but can still concentrate on his job. Dual pathways of compression performing in harmony Modern digital signal processors in hearing instruments manage and control hundreds of parameters. In many cases these affect each other, sometimes with negative results. What distinguishes an effective signal processor is the art of harmonizing hundreds of different parameters to avoid artifacts and internal interference signals. Even in such a well developed area of signal processing as compression, attention to detail is paramount in providing the wearer with the best clarity and comfort. The need to react quickly and accurately to different situations means that compression systems often employ a dual-path compression strategy, with one path providing slow-acting, the other path providing fast-acting gain control. However, the fast-acting path with its short time constants can blur the amplitude fluctuations of the original signal and the slow-acting path can impede responsiveness in managing sudden uncomfortably loud sounds. Some competitive systems use a process whereby both compression paths are directly controlled by averaging the input signal over time. Since an average is used to determine compression settings, neither may be ideal. Returning once more to opening night at the city concert hall, consider the moment that the performance actually begins. The lights dim and the audience chit chat fades to silence. The gaffer s spotlight pierces the darkness to reveal the orchestra poised for the conductor to bring them to life. With one fell swoop of the baton, the orchestra explodes in perfect harmony flooding the concert hall with the opening chord. At this moment, the gaffer is vulnerable to this sudden acoustic onslaught. In this scenario, a time-input average control system would first increase the gain, causing a substantial overshoot, before finally applying the appropriate compression as shown by the grey line in Figure 4. Figure 4 - Adjustments to gain over time in response to an abrupt increase in signal intensity Furthermore this unique adaptive dual-path compression system is integrated into the SoundFlow automatic system, so amplification control is both time and situation-sensitive. As appropriate for the situation, the time constants which provide best signal transmission with the least distortion are automatically selected. This achieves instantaneous and smooth automatic adjustments, so even the most abrupt changes to the sound environment are efficiently managed without distortion or artifacts. Directionality is a Phonak passion Phonak has always led the way in directional microphone technology. The Spice platform once again takes the industry benchmark one giant leap forward with two pioneering innovations in directionality. Directional noise cancelling goes spatial Traditionally, regardless of microphone mode, the noise cancellation algorithm is always applied in the same manner based on temporal cues. However when in directional mode, the source of interest is actually known, so a new approach can be applied UltraZoom with SNR-Boost is the next revolution in adaptive multi-channel beamforming technology. SNR-boost is a spatial noise canceller used in conjunction with UltraZoom, the Phonak beamformer. In contrast to the temporal-based noise canceller, NoiseBlock, it places focus on the direction of the incoming sound as shown in Figure 5.

17 Traditional directionality UltraZoom with SNR-Boost Figure 5 - Traditional directional microphones amplify everything within in the beam including unwanted noises. UltraZoom with SNR-Boost effectively cancels noise, enhancing SNR for speech coming from the front NoiseBlock works by analysing variations of sound over time across multiple frequency bands, reducing gain for different frequency bands whenever it detects noise. In contrast, SNR- Boost analyses the direction-of-arrival of the sounds and if a target sound from the front is detected, reduces gain for sounds coming from the back hemisphere. In contrast to most noise cancellers it keeps target sounds from the front more accurately and reduces noise from behind much more selectively. Even when the noise contains speech. Of course, SNR-boost also acts individually in different frequency bands and applies gain reduction tailored to the conditions of the current sound. In combination with NoiseBlock, and dynamically activated from Soundflow, both algorithms act in unison and are fine tuned to their individual strengths. Does a hearing system beat a pair of hearing instruments? The answer with the Phonak Spice Generation is a clear yes. By cleverly combining a binaural directional advantage with sophisticated wireless and real broadband audio exchange capabilities, a hearing industry first has been created. A single omni-directional microphone captures sound from all around, adding a second microphone creates some directionality. Acoustic theory says that the more microphones that are available, the greater the level of directionality (Brandstein and Ward, 2001) 3. However there is a problem in adding more microphones to a hearing instrument as this would make it bigger. So there was a dilemma for the Phonak engineers, how to add another microphone without making the instrument bigger. Phonak engineers are highly adept at thinking outside the box. This led them to the realization that another microphone was already available for a binaural wearer - the instrument on their other ear. Not only were there 2 extra microphones but the relatively large distance between the instruments contributed significantly to the low-frequency directionality as shown in Figure 6. Figure 6 Directivity Index improvement with StereoZoom This multi-microphone configuration by itself was only half the story. In order for the contralateral microphones to contribute to directionality, the full audio signal from both devices must be available for integrated processing. Thankfully Phonak had the foresight to introduce this capability on its previous generation platform - CORE. This capability allowed binaural features such as ZoomControl and DuoPhone in CORE products. With Spice processing power, StereoZoom has become possible providing improved speech intelligibility and reduced listening effort as evidenced in clinical trials (Phonak AG, 2010) 4. StereoZoom is a classic example of the whole being greater than the sum of the parts. Summary The Phonak Spice generation is testament to what is possible with a holistic approach to innovation. With one part of this platform, the Spice chipset, multitudes of signal processing features are orchestrated in harmony, realizing an unparalleled hearing experience for wearers. This Phonak Insight has presented just a taste of what Spice makes possible. References 1 Nyffeler M: Software seeks to provide seamless adaptation to changing soundscapes. Hear J 2009:62(10): Phonak AG: FlexControl Individualizing automatic performance. Field Study News Sept Brandstein M, and Ward D Microphone Arrays Signal Processing Techniques. Berlin: Springer-Verlag 4 Phonak AG: StereoZoom Improvements with directional microphones. Field Study News Sept 2010

18 Field Study News Spice+ Processing Superior first-fit acceptance Abstract The Phonak fitting philosophy is to provide a pleasant sound quality while also ensuring optimal audibility and intelligibility. The new Phonak Spice+ product portfolio launched in October 2011 feature a new signal processing algorithm called Spice+ Processing. The innovation focus for Spice+ Processing was to increase spontaneous user acceptance and provide the most natural hearing experience in quiet. This Field Study News reports on two studies conducted to investigate the achievement of these objectives. Introduction In the hearing instrument fitting process, hearing care professionals seek a positive initial reaction from the client without too much modification of the manufacturer's first-fit algorithm (Van Vliet, 2009). However, the precalculated targetamplification does not always meet the desired loudness and sound impression for the hearing impaired. This can be particularly the case for first time hearing aid wearers who often do not accept hearing aid settings without a corresponding phase of acclimatization (Cox et al., 1996; Munro and Lutman 2004). In fact some researchers go so far as to suggest that measuring initial satisfaction with hearing devices should be postponed by one month because selfreporting, often used in hearing aid fitting assessments, may be meaningless without any real-life experience (Kuk et al, 2003). All this speaks for an optimal first-fit precalculation to ensure the first impression with amplification or a new hearing instrument is as positive as possible. A common approach is to sacrifice audibility and intelligibility for the sake of immediate, short-term acceptance. In contract, the Phonak fitting philosophy has always been to offer a pleasing overall sound experience while fostering the best possible speech intelligibility. The Phonak Spice+ product range, launched in late 2011, features Spice+ Processing, a new signal processing algorithm designed with two key benefits in mind: increased spontaneous acceptance and a more natural hearing experience in quiet. Increased spontaneous acceptance The trend towards more open fittings (Johnson, 2008; Kochkin, 2011) for mild to moderate hearing losses means a fitting precalculation needs to take into account not only the amount of amplification lost through venting, but also the level of sounds presented directly to the eardrum, when calculating the appropriate amount of gain to compensate for the hearing loss. This interaction, between unamplified signals coming through the vent and the amplified sound coming from the instrument, must be accurately accounted for when calculating the most appropriate gain prescribed as part of a first fit precalculation. Spice+ Processing includes an enhanced fitting formula which better addresses the variability from the individual properties of the ear, the chosen acoustic coupling, the direct sound reaching the ear and amplified sound. More natural hearing in quiet Quiet listening situations actually change more than is realized. For example, a quiet situation may be a soft conversation with or without low level background noise or complete silence. Spice+ Processing includes a refined expansion or soft-squelch approach which effectively manages subtle signal amplitude variability in real time. It also utilizes adaptive time constants to address temporal fluctuations in input signals. The design aim of these changes to the processing algorithm was to significantly enhance spontaneous acceptance, reduce fine-tuning effort for the hearing care professional and maintain optimum audibility and speech understanding. To validate these Spice+ Processing objectives, two studies were conducted. The purpose of the first study was to test the subjective initial reaction to the first-fit settings using two different hearing instrument models, with subjects with mild to moderate hearing losses. The second study compared the Phonak Ambra M H2O with Spice+ Processing against various other hearing instruments to investigate sound quality ratings and spontaneous acceptance using paired comparisons.

19 Study 1: Method The Spice+ Processing validation study conducted in Switzerland included 20 test subjects wearing micro-size BTE and 10 wearing miniature CIC devices. Of the 20 micro-size BTE test subjects, nine had either no or less than 6 months wearing experience with hearing instruments, three between six months and three years, four between three and six years and four test subjects had over six years of experience with hearing instruments. The average age was 61 years (min. 48 and max. 73 years), with 3 female and 17 male patients. The average hearing loss for this group is shown in Figure 1. Procedure Several subjective measurements were carried out during home trials and lab tests using questionnaires to investigate the spontaneous acceptance of the new Spice+ precalculation. The lab tests included presentations of three different sound samples (birdsong, female speech and water) at 55 db. Clients rated loudness and sound quality. Additionally observation protocols (filled out by the fitters) noted spontaneous test subject comments regarding loudness, sound quality as well as the fitter fine-tuning effort required. Objective testing included evaluation of speech intelligibility in quiet using the Freiburger word recognition test and in noise using the OLSA Oldenburg Sentence test (Wagener et al, 1999). Study 1: Results Subjective ratings of the sound quality and loudness of both the test subjects and fitter s voice are shown in Figure 3 below. Fig.1: The average hearing loss and standard deviation of the 20 micro BTE validation participants. Of the 10 miniature CIC test subjects, six had no wearing experience with hearing instruments, one between six months and three years, and three test subjects had over six years of experience with hearing instruments. For this group, the average age was 66 years (min. 41 and max. 77 years). All test subjects had good manual dexterity. The gender mix was balanced with 5 female and 5 male participants. Figure 2 shows the average hearing loss for this group of test subjects. Fig. 3: Sound quality and loudness ratings with Spice+ micro-size BTE Loudness of both own voice and the fitter s voice was rated as just right, with only 5% of the test subjects rating the fitter s voice as soft but comfortable. Of the test subjects, 65% rated their own voice as good, and 35% as not good. These were nearly all first time hearing aid wearers. The Phonak nano test subjects group results showed the following sound quality and loudness ratings (Fig. 4). Fig.2: The average hearing loss and standard deviation of the 10 miniature CIC validation participants Devices The micro BTE group was fitted with Phonak Ambra M H2O instruments while the miniature CIC group was fitted with Phonak Ambra nano devices. All test subjects were fitted binaurally. Fig. 4: Sound quality and loudness ratings with Spice+ nano device

20 The fitters also asked the test subjects if they felt any hearing instrument fine tuning was required (Fig. 5). For both groups this was asked after performing an first-fit precalculation based on the test subject s hearing loss, acoustic coupling and hearing aid experience. Fig. 5: Fine tuning required after precalculation for the micro-size BTE and nano instruments. A total of 7 out of 30 test subjects required some degree of fine-tuning after precalculation, predominantly for first time hearing aid wearers with own voice (loudness or sound quality) issues. Study 2: Method The explorative study at the Hörzentrum in Oldenburg was carried out to conduct paired comparisons between the Spice+ Processing precalculation as reference and various Phonak and competitive products and precalculations. The aim of the study was to investigate subjective ratings regarding sound quality and speech understanding or clarity with a variety of sound samples. The 15 test subjects, aged between 31 and 76 years old, had an average age of 66.2 years. The group comprised of 13 male and two female patients, eight of which had previous experience with hearing instruments. Devices All 15 test subjects were fitted with Phonak Ambra M H2O (Spice+), Phonak Ambra microp (Spice) as well as with two competitive state-of-the-art hearing instruments (Competitor A and Competitor B). Procedure The settings of all hearing aids were pre-calculated by the respective fitting software based on the audiogram and acoustic coupling information. Where possible, settings were configured using the patient age, gender, hearing aid experience and type of previous processing. Feedback thresholds were determined and the feedback cancellation functionality was activated. The hearing instrument experience level tool was then used to achieve good initial acceptance and no further fine-tuning was applied. At this stage the test subjects spontaneous comments and sound quality ratings were noted. Individual ear canal recordings with the hearing instruments insitu and selected sound samples were then made and saved into a virtual hearing aid. On the test subjects second visit paired comparison ratings using the virtual hearing aid were conducted. The paired comparison sound samples included soft speech (46 db A) in quiet, moderate speech (56 db A) in quiet and speech in noise with SNRs of -5 db and -10 db. The recordings were made with the target speech signal at 0 and where applicable interferer noise at 45, 90, 135, 180, 225, 370 and 315. The hearing instruments were set to the appropriate quiet or noise program depending on the sound sample being recorded. The order of the paired comparisons was randomized and test subjects were asked to provide ratings on loudness, clarity, sound quality and intelligibility. The average hearing loss can be seen in Figure 6. Fig. 6: The average hearing loss and standard deviation of the Oldenburg validation participants.

21 Study 2: Results As one of the design aims of the Spice+ Processing was to optimize sound quality for quiet listening situations, the subjective ratings of sound quality in total and compared to the previous generation Spice instruments was of particular interest. Figure 7 below shows the results when rating speech in quiet. Total Spice Total Spice Competitor A Competitor B Competitor A Competitor B Fig. 8: Subjective ratings, total and by instrument, to the question with which hearing instrument do you understand better with moderate speech in quiet. Fig. 7: Subjective ratings, total and by instrument, to the question which hearing instrument has a better sound quality with moderate speech in quiet. The reference instrument was always the Spice+ instrument. The green area shows the percentage of test subjects who preferred the Spice+ hearing instrument compared to the other comparison hearing instruments. When comparing Spice+ to all comparison hearing instruments, 49% of test subjects preferred Spice+ and 22% had no preference. When comparing the Spice+ instrument to the Spice instrument, 80% either preferred the Spice+ instrument or had no preference, while just 20% preferred the previous generation Spice. There was no difference between Spice and Spice+ and a strong preference for Spice+ when comparing it to the Competitor A hearing instrument. In the case of competitor B, for speech in quiet, the test subject showed no clear preference. This was rather different when rating sound quality, as seen in Figure 7. Figure 9 below shows the total results for the question with which hearing instrument do you understand better. The measurement condition in this case was speech in noise with an SNR of -5 db and -10 db. Total, -5 db SNR Total, -10 db SNR The Spice+ Processing algorithm aims to ensure not only enhanced spontaneous acceptance and sound quality in quiet, it also has to ensure optimal speech understanding both in quiet and in noise. As Figure 8 shows, the subject s personal preferences in terms of speech understanding when listening to speech in quiet at a normal level showed an overall preference for Spice+. Fig. 9: Subjective total ratings to the question with which hearing instrument do you understand better at -5 db SNR (top) and -10 db SNR (bottom).

22 This showed that in comparison to Spice+, the preference for the previous generation Spice and competitor A and B instruments was only 20% and 16% respectively. The percentage of test subjects who had no preference was also relatively high, 29% and 42%. Looking at the data in detail, this was due to the high no preference rating between Spice and Spice+, as shown in Figure 10. Spice Competitor A Competitor B Spice Competitor A Competitor B Conclusion The overall conclusion from the first validation study shows a very high subjective rating of the Spice+ Processing first-fit settings. Own voice and fitter voice loudness ratings was rated just right between 90 and 100% for the two groups, as was the fitter s voice sound quality rating. Very little fine-tuning after precalculation was needed and if so, this was mainly for the first time hearing instrument wearers who required a little fine tuning for the sound of their own voices. This study shows a very high spontaneous acceptance of the Spice+ Processing precalculation and initial settings. The second study also showed very good subjective ratings of the Spice+ Processing hearing instrument. The results showed that ratings of speech understanding and sound quality in quiet were highest for Spice+ compared to the other, previous generation and state-of-the-art competitive instruments. The results also show the Spice+ instrument ranked highest in terms of speech understanding in a relatively easy situations as well as in a difficult speech in noise listening situation, when compared to two competitive hearing instruments. References Fig. 10: Subjective ratings, by instrument, to the question with which hearing instrument do you understand better at -5 db SNR (top) and -10 db SNR (bottom). From Figure 10 we can see that the preference to the comparison devices was between 13% and 33%. Spice+ was preferred over both competitor A and B instruments in the speech in noise listening situations. Cox, R. M., G. C. Alexander, I. M. Taylor and G. A. Gray (1996). Benefit acclimatization in elderly hearing aid user. J Am Acad Audiol, 78(6), Johnson E. (2008) Practitioners give high marks for user benefit to open-canal mini-btes. Hear Jour, 61(3), Kochkin S. (2011) MarkeTrak VIII: Mini-BTEs tap new market, users more satisfied. Hear Jour, 64(3), Kuk, F.K., Potts, L., Valente, M., Lee, L. & Picirrillo, J. (2003) Evidence of acclimatization in persons with severe-to-profound hearing loss. J Am Acad Audiol, 14, Munro, K.J. & Lutman, M.E. (2004) Self-reported outcome in new hearing aid users over a 24-week post-fitting period. Int J Audiol, 43, Van Vliet, D. (2009) Final Word: Who s in Charge? Hear Jour; 62(5) Wagener K, Kuehnel V, Kollmeier B. (1999) Development and evaluation of a German sentence test; Part I-III: Design, Optimization and Evaluation of the Oldenburg sentence test. Zeitschrift für Audiologie. 38:86-95 For more information, please contact: audiology@phonak.com

23 Field Study News Power SlimTube: Cosmetically attractive fit without losing performance Summary The new Power SlimTubes now offer Naída S wearers the benefits from advances in acoustic coupling design without a significant loss in gain and output. The goal of this study was to investigate if Power SlimTubes provide a more cosmetically attractive solution on the ear than the standard tubing, but without any changes to the audiological performance. The results of objective measurements of the Naída S with Power SlimTube or standard tube as acoustic coupling show that the speech intelligibility and output on the ear were not significantly influenced by tubing type. Introduction The launch of the slim tube has brought a cosmetic revolution for hearing impaired people with a mild to moderate hearing loss. The cosmetic improvements achieved were based on the most visible part of a BTE from the front, the tone hook and tubing. Mueller (2006) described that, in addition to audiological performance, cosmetic appearance and comfort of an open-fit hearing system are the most relevant factors in the client s purchasing decision. Power wearers, clients with a severe to profound hearing loss, have previously been excluded from improved cosmetics because using a slim tube results in a significant reduction in gain and output, impacting audibility. But power wearers also take the cosmetic appearance of a hearing system into consideration during the purchasing decision. With the new Power SlimTube Naída S wearers can now benefit from advances in acoustic coupling design. The Power SlimTube, offered with a standard earmold, provides all the cosmetic advantages while minimizing the loss of gain and output. In this study, the results of the objective real ear measurements and the speech tests in quiet and noise shows little or no changes in audiological performance for Power SlimTubes compared to a tone hook and standard tubing. Goal of the Trial The aim of this study was to investigate if the fit of the Power SlimTubes are cosmetically more attractive on the ear than a tone hook and standard tube. The cosmetic benefit of Power SlimTube on the ear should not influence the audiological performance when compared to the known acoustic coupling of the standard tube. Set-up of the Study All subjects received two pairs of similar earmolds, one pair with the standard tube and one pair with Power SlimTube as the acoustic coupling. The hearing aid setting was the same. The only differences were the settings of the correct acoustic coupling in the software (standard tube or Power SlimTube including the length), and the measured feedback test for both acoustic couplings, which affected the hearing aid precalculation. The following objective measurements were done to compare the Power SlimTube with the standard tube: 1. Real ear aided response measurements were done with the Siemens Unity measurement equipment. An International Speech Test Signal (ISTS) of 65 db was used for the measurement. 2. Speech test in quiet with monosyllabic words were presented from a loudspeaker in front of the subject at 1m in distance. The speech level was 50, 65 and 80 db for the measurements. The results were expressed in percentage of correct discrimination. 3. An adaptive speech test in noise with an uncorrelated background noise at 65 db was measured. The speech signal was presented at starting level of 65 db in front of the subject. The uncorrelated background noise was presented from 5 loudspeakers around the subject. The subject was sitting in the circle of the loudspeakers at a distance of 1.4m from all loudspeakers. The results were expressed as SNR (speech to noise ratio) in db. Additionally, for all subjects photographs were taken to show the fit on the ear using a Naída S with a tone hook and standard tube and a Naída S with a Power SlimTube. The subjects answered in a questionnaire if they rated the Power SlimTube more cosmetically attractive as the standard tube, and which acoustic coupling they preferred after first week wearing time.

24 Subject and Devices 18 subjects (16 male and 2 female) with severe to profound hearing loss took part in the study. Half of the subjects were fitted with Naída S SP and the other half with Naída S UP hearing aids. All subjects had two pairs of similar earmolds, one pair with the standard tube and one pair with the Power SlimTube as the acoustic coupling. Speech intelligibility in noise was not significantly different (p=0.925; Wilcoxon matched pairs test) for both acoustic couplings. Figure 3 show the averaged data for both acoustic couplings for n=14 subjects with Naída S SP and UP. 4 subjects with a profound hearing loss could not finish the speech in noise test due to an exclusion criteria of SNR > +15dB. The speed of the speech presentation together with the background noise was too difficult for those 4 subjects. Results Picture 1 and 2 show the fit of the standard tube compared to the Power SlimTube on the ear for one subject. The results from the questionnaire show that all subjects rated the Power SlimTube spontaneously as cosmetically more attractive. After one week of wearing time, 67% of the subjects preferred the Power SlimTube. Figure2: Speech test in quiet measured at 50, 65 and 80 db speech level with Power SlimTube compared to standard tube coupling. Pic. 1: Naída S with standard tube Pic.2: Naída S with Power SlimTube The averaged curves of real ear aided response measurement (figure 1) show nearly the same amplification on the ear for Naída S with both acoustic couplings. Figure3: Speech test in noise measured with Power SlimTube compared to standard tube coupling. Conclusion The outcome of this study shows that Power SlimTubes are not only cosmetically more attractive on the ear compared to standard tubing and tone hook but objective measurements are nearly the same results for both acoustic couplings. The cosmetic advances of the Power SlimTube for Naída S are a feasible option for power wearers without affecting gain or speech intelligibility. References Figure 1: Real ear aided response measured with ISTS at 65 db for Naída S with Power SlimTube compared to standard tube for n=18. Figure 2 shows the results of the speech test in quiet for Power SlimTube compared to standard tube coupling. There was no significant difference between both acoustic couplings. Dechant, 2009, Hörakustik: Kabelbiegung fast wie magefertigt 10/2009, Mueller, 2006, Hearing Journal: Open-Canal Fittings: A Special Issue. 59 (11), For further information please contact: davina.omisore@phonak.com

25 Field Study News Phonak CROS: A quantum leap for people with total unilateral hearing loss Abstract For people with total unilateral hearing loss, complete loss of hearing in one ear and better hearing on the other ear, understanding speech in noisy situations can be a great challenge which is exacerbated by the head shadow effect. There are many systems available which transmit the acoustic information from the unaidable side to the hearing side but most have not been very efficient. A CROS system consists of a CROS transmitter microphone on the unaidable side and a hearing instrument receiver on the side with better hearing. The new Phonak CROS System, based on the new Spice Generation platform and its wireless capability (HiBAN), makes it possible to transmit clear acoustic information without the intrusion of cable connections between transmitter and receiver. Phonak CROS reduces the head shadow effect enabling users to better understand speech from the unaidable side in noisy situations. 20 test subjects took part in a study to investigate speech intelligibility in noise for people with total unilateral hearing loss wearing Phonak CROS. Six of the subjects were experienced CROS users. Objectively recorded measurement results showed a clear improvement in speech intelligibility in noise with Phonak CROS. The subjective data were obtained by means of a client questionnaire, and likewise reflected a high level of satisfaction amongst the test subjects. Introduction The appeal of Phonak CROS lies in robust wireless audio signal transmission from the unaidable ear to the better ear, its esthetically pleasing design, and its ease of use. One great advantage of Phonak CROS, the wireless HiBAN network, makes Phonak CROS compatible with all wireless Phonak Spice hearing instruments. It was important that Phonak CROS overcome the limitations of current CROS systems in terms of size and shape, so that it satisfies the requirements of each individual. People with normal hearing in the better ear (CROS) as well as people with mild to profound hearing loss (BiCROS) will benefit from the wide variety of functions Phonak CROS offers such as SoundFlow, Real Ear Sound, QuickSync and, for BiCROS fittings only, SoundRecover. This makes Phonak CROS a genuine highlight amongst today s CROS systems. The new Phonak CROS is available as a BTE transmitter in the Audéo S SMART housing. It can be placed on the ear either with the specially designed Phonak CROS Retention, or with the customized Phonak CROS Tip to comfortably fit the individual contours of the ear. The CROS transmitter is also available as an custom shell option (ITC, half-shell and fullshell). The different CROS transmitter styles are shown in Figure 1. Fig. 1: Illustration of the Phonak CROS BTE and ITE transmitter options. Due to its high degree of flexibility, Phonak CROS can be combined with any type of Spice hearing instruments. Aim of the study The aim was to determine whether there is a clear improvement in speech intelligibility in noise for people with a total unilateral hearing loss when using Phonak CROS. Method To determine the speech intelligibility, the Oldenburg sentence test (OLSA) was carried out. The speech intelligibility was ascertained through adaptive measurement of the speech reception threshold (SRT, signal-to-noise ratio (SNR) at 50% speech intelligibility). The noise used was a speech-

26 simulating noise. The test subject was rotated in a circle of 12 loudspeakers, once in the direction of the 90 or 270 loudspeaker (seating position 1; simulating conversation partner at the side), and then in the direction of the 60 or 300 loudspeaker (seating position 2; simulating conversation partner obliquely in front), so that in each case he was seated with the unaidable side toward the 0 loudspeaker. The speech signal produced was always presented from 0. The noise was played for seating position 1 from 60, 120, 180, 240 and 300 and for seating position 2 from 60 or 300, 120, 180 and 240. The measurements were taken using the Speech in Noise hearing program, with the BTE CROS transmitter set to the microphone mode Real Ear Sound, and the ITE CROS transmitter set to the Omnidirectional microphone mode. For the BTE measurements, the subjects wore Audéo S SMART IX hearing instruments in the better ear. The OLSA test was carried out with CROS subjects with and without the Phonak CROS System (Phonak CROS plus a Spice hearing instrument). The BiCROS were taken with just the BTE receiver and then with the complete system. The subjective data were collected by means of client questionnaires, which were completed at home during the test phase. Fig. 2: The averaged measurement data (N = 9) show a clear improvement in the SNR with the Phonak CROS BTE system for both seating positions. The lower the measured values the better the speech intelligibility result. The additional benefit of the Phonak CROS transmitter to wearing a receiver hearing instrument can be clearly demonstrated. The measurement results for the BiCROS subjects fitted with ITE devices (Fig. 3) also show the improvements similar the BTE devices. Seating position 1 showed a SNR improvement of 4.7 db and a 5 db SNR improvement for seating position 2. Test subjects and hearing systems A total of 20 test subjects took part in this study, six of whom were already experienced users of CROS devices. Both BTE and ITE CROS systems were tested. The first part to the validation investigated the Phonak CROS BTE. 14 subjects were fitted with Audéo S SMART IX to their better ear and five subjects had Phonak Cassia BTE or Phonak Solana BTE fitted to their better ear. After completion of the BTE validation, six BiCROS subjects were fitted Phonak CROS ITE transmitter and Phonak Ambra 312 UZ ITE as the receiver instrument. All the subjects tested the Phonak CROS System in the laboratory and in everyday life situations. Results The averaged measurement results from the OLSA test showed a clear improvement in the intelligibility of speech in noise with the Phonak CROS System. For seating position 1, a SNR improvement of 3.4 db was achieved with the Phonak CROS System, and for seating position 2 an improvement of 2.5 db was achieved in comparison to only the receiver microphone situation. It was possible to demonstrate that both CROS and BiCROS subjects clearly benefited from Phonak CROS. The results presented here relate to the BiCROS subjects fitted with BTE and ITE devices. The measurement results for the BTE BiCROS are shown in Figure 2. Fig. 3: The averaged measurement data for the ITE BiCROS subjects (N = 6) show a clear improvement in speech intelligibility with the Phonak CROS system in both seating positions. The lower the measured values the better the speech intelligibility result. For the CROS test subjects, good results were similarly achieved for the same seating positions (data not shown). The effect of the head shadow can thus be attenuated through the use of Phonak CROS, meaning that improved speech intelligibility from the unaidable side in noisy situations is once again possible. It is well known that the head shadow effect influences not only the intelligibility of speech in noisy situations, but also the sound that is perceived. A brighter sound impression arises from the unaidable side, since high frequencies, which are harder to bend around the head, are now compensated with the CROS system. The subjects describe the altered sound impression as a way to determine the location of sound sources, and did not find it irritating. This is clearly an additional benefit of the Phonak CROS transmitter over a conventional unilateral hearing instrument solution. In the client questionnaires, the sound quality of the Phonak CROS System was rated as pleasant and natural by 80% of all subjects. This illustrates the good sound quality of Phonak CROS.

27 Conclusions The new Phonak CROS System clearly demonstrated that people with total unilateral hearing loss will perceive better speech understanding in noisy situations from the unaidable side. This is possible through the stable, wireless broadband audio signal transmission from Phonak CROS to the Spice hearing instrument receiver, in real time, with outstanding sound quality. Phonak CROS is a quantum leap in the treatment of total unilateral hearing loss, with the most cosmetically appealing design since the invention of the CROS systems. References H. Ericson et al. (1988), Contralateral Routing of Signals in unilateral hearing impairment A better method of fitting, Scand. Audiol. 17, p: For further information, please contact: Carmen.Steitz@phonak.com

28 Field Study News AudiogramDirect: In-situ hearing tests at their best Summary The latest version of the Phonak fitting software, Phonak Target 1.2, sees the return of AudiogramDirect. This in-situ hearing test enables fitters to check client s hearing directly through any Spice hearing instrument, taking into account the properties of the individual ears and the chosen hearing instruments with their acoustic coupling. This method provides a fast and accurate point for a successful fitting when used in conjunction with a diagnostic hearing test. With AudiogramDirect you can accurately fit any Spice hearing device directly from your laptop or PC, without the use of any additional audiometric equipment. A validation study, involving 39 participants with various degrees of hearing loss fitted with a selection of devices, investigated the reliability of AudiogramDirect compared to standard diagnostic audiometry measured with the Aurical system using headphones. The results conclude that AudiogramDirect is a good way to tests clients hearing from mild to profound hearing loss. Introduction Most modern digital hearing aids now have on-board sound generators that produce frequency specific pure tones, so an in-situ hearing test can be performed. The ability to test the hearing thresholds directly through the hearing aid placed in the ear (in-situ), makes the fitting more precise, achieving an accurate and custom initial fitting that will have a huge impact on initial satisfaction and the overall success of a professional practice (Block, 2008). In-situ hearing tests can also be used to observe how standardized audiometric hearing levels will vary because of the influence of residual ear canal volume (Keidser et al, 2011). The procedure also takes into account the effects of the depth of the instrument in the ear canal, the effectiveness of the acoustic coupling seal in the ear canal, the effects of venting, and the specific receiver in that instrument (Block, 2008). This correction allows the target gains to represent the hearing loss more accurately (Keidser et al, 2011). The real-ear measures enable the hearing aid to match those target gains with more precision (Block, 2008). The result is a fitting that is based on the actual characteristics of your client ear rather than average data. This makes it possible to customize the fitting responses for your specific patient and, by doing so, improve the accuracy of the fitting (Block, 2008). In-situ audiometry is an attractive option because it requires less equipment and resources, and may save on clinical time used when transferring threshold data between different test modules. One downfall is that in-situ audiometry is currently limited to measurements of air conduction thresholds (Keidser et al, 2011). Another disadvantage of in-situ audiometry is that it requires special equipment like the hearing aids before the hearing test can be performed. It also cannot be used to directly compare with other measurement devices in the market. Finally, in-situ hearing tests should be used to compliment already measured diagnostic audiometry and not as a measurement on its own. Goal of the Trial The goal of this validation study was to determine if audiograms tested with AudiogramDirect and a Spice hearing instrument are comparable and reliable as measurements made with traditional audiometry (TA) with headphones. Studies have suggested that both behavioral and/or physiological changes can lead in a test-retest variability of audiometric test results (Stuart et al, 1991; Landry et al, 1999) of up to 10-15dB. The study assessed if measured points using AudiogramDirect fall into a tolerance range of +/- 10 db for mild to moderate hearing loss and moderately severe to profound hearing losses for the single frequencies. To increase reliability of findings the test was performed with different acoustical coupling options: xs receivers with open dome; xp receiver with closed dome; SlimTubes with open and closed domes and individual ear pieces. The reference measures were performed with a standard audiometer, Aurical, in the traditional way.

29 Subject and Devices Test subjects A total of 39 participants took part in the validation study. There were 19 participants, 3 female and 16 male, with a mild to moderate hearing loss, average age was 68 years (fig.1). The remaining 20 participants, 18 male and 2 female, had a moderately severe to profound hearing loss, with the average age of year (fig.1). Results All the results were reliable and robust, falling into the defined range of +/- 10 db for mild to profound hearing losses. Mild to moderate hearing loss 38 Aurical audiograms (left and right, 7 main frequencies: 250Hz, 500Hz, 1kHz, 2kHz, 3kHz 4kHz and 6kHz) were compared to AudiogramDirect measures. 94% of all AudiogramDirect measurement points with Audéo S IX and xs (right ear) or xp (left ear) CRT were within the predefined +/- 10 db range (fig. 2). 89% of all AudiogramDirect measurement points with Ambra microm with SlimTubes were also within in the predefined +/- 10 db range (fig. 3). 6.8% of the points fell out of the defined range. This could be related to concentration issues or environmental noise effects and therefore can be neglected. The variance doesn t show any specific direction. Figure 1: Averaged hearing loss of all particiapnts Devices To increase reliability of the AudiogramDirect findings as well as to cover a broad range of product styles and acoustical coupling, the following set-up was chosen for the participants with a mild to moderate hearing loss. The participants were fitted with Audéo S SMART IX CRT devices and Ambra microm devices. Audéo S SMART IX devices fitted on the right ear had xstandard (xs) receivers and open domes while devices fitted on the left ear had xpower (xp) receivers and closed domes. Ambra microm devices fitted on the right ear had SlimTubes with open domes and the left ear fitted with SlimTubes and closed domes. The participants with a moderately severe to profound hearing loss were fitted with Naída S IX SP and UP devices and individual earpieces. Figure 2: All AudiogramDirect measurement points (250Hz, 500Hz, 1kHz, 2kHz, 3kHz, 4kHz and 6kHz) for Audéo S IX in comparison to traditionally measured audiogram (TA) with Aurical with a +/- 10 db range. Test Method The participant s standard audiogram was measured using the Aurical system and headphones approximately 1.5 months prior to measuring with AudiogramDirect. Phonak Target 1.2 with AudiogramDirect was used to measure the participants hearing using the Spice hearing aids and the appropriate acoustic parameters for the participants hearing loss. Figure 3: All AudiogramDirect measurement points (250Hz, 500Hz, 1000Hz, 2000Hz, 4000Hz and 6000Hz) for Phonak Ambra in comparison to traditionally measured audiogram (TA) with Aurical with a +/- 10 db range.

30 Moderately severe to profound hearing loss For the moderately severe to profound hearing loss, AudiogramDirect measures compared to Aurical audiograms showed good results and did not deviate by more than 10 db for 94.1% of all AudiogramDirect measurement points (fig. 4). In the lower frequencies, some clients showed a larger deviation which could be due to the vent leakage of the ear pieces. The ear pieces of some clients did not fit well in the ear canal, so the real vent effect could be larger than the chosen one in the acoustic coupling. In the high frequencies there were less measurement points because most profound hearing losses were out of the measurement range of AudiogramDirect which limits measuring up to 6 khz and not beyond 100dB. References For further information please contact Davina.omisore@phonak.com Figure 4: AudiogramDirect results for Naída S IX with individual earmolds to traditionally measured audiogram (TA) with Aurical with a +/- 10 db range. Conclusion The outcome of this study shows that when used in conjunction with a diagnostic hearing test already performed, Audiogram Direct is a quick, reliable and robust tool to check clients hearing during follow ups or if there is a change in their hearing. The results for clients with a mild to moderate hearing loss with CRT or BTE devices and a variety of acoustic couplings show that AudiogramDirect can be used as a reliable hearing test tool with a wide range of hearing loss. For moderately severe to profound hearing losses, AudiogramDirect also performs as a very useful tool enabling hearing care fitters to give their clients a more individual and hearing aid specific fitting.

31 Field Study News SoundFlow: Seamless adaptation to every soundscape

32 No. of Transitions Perceived Adaptations with AutoPilot and SoundFlow Subject AutoPilot SoundFlow Paired Comparison "Speech intelligibility" 35% 21% 44% SoundFlow clearer AutoPilot clearer No Difference

33 Field Study News WhistleBlock Technology: The new benchmark in feedback elimination WhistleBlock Technology The new benchmark in feedback elimination For decades, feedback, whistling or squealing, has been a major complaint of hearing instrument users [Kochkin, MarkTrak I-IV ]. The underlying physical mechanism of feedback is rather simple: the amplified sound from the receiver / loudspeaker leaks through the vent and is picked up and re-amplified by the hearing instrument. This causes acoustic instability which ultimately leads to a well-known and very annoying whistling, squealing or howling in the hearing aid. In the past, the major approach to handle feedback was to either limit the applied gain or reduce acoustic leakage with small vents. The introduction of digital technology has significantly improved the acoustic stability of modern hearing instruments. Modern feedback management systems enable hearing care professionals to optimally use the residual dynamic range of the individual hearing impaired person while also using acoustic coupling systems with larger vent diameters. In particular, the introduction of open fitting devices offering a much higher wearing comfort has been made possible thanks to modern feedback management systems. Despite significant improvements in the acoustic stability of hearing instruments, the performance of today s feedback management systems is still mainly driven by trading off feedback cancelling performance with sound quality and effectively applied gain. A further challenge of state-of-the-art feedback management systems is that they can cause distortions of natural signals, such as music, telephone ringing or door bells. Physical mechanism challenges for feedback management systems As mentioned above, the underlying basic mechanism of feedback is rather simple. However, it is somewhat more complex when it comes to the parameters influencing the occurrence of feedback. The feedback transfer function, which is determined by the acoustic feedback path between receiver and microphone, is not stable but changes significantly during the course of the day. This is due to the fact that the wearer might move an object close to the hearing instruments (i.e. telephone), might walk along or sit next to walls or objects, wear hats, talk and yawn. [J. Hellgren (1999)] did a systematic analysis of the different parameters influencing the generation of feedback. The major conclusions from his studies were: There are large differences in terms of spectral, temporal and amplitude characteristics for different feedback generation mechanisms. Rather large differences between various subjects are observed due to different ear canal and pinna anatomies. Feedback is not a phenomenon occurring at a single frequency. It has complex, time varying spectral characteristics, but it is typically most prominent in the spectral range around khz. Overall, feedback is a complex, highly dynamic phenomenon requiring complex adaptive feedback path estimation and cancellation techniques to be tackled. Besides the dynamic changes in the acoustic transfer path, other algorithms running in the hearing instruments such as the dynamic, compressive amplification schemes or adaptive noise reduction systems must also be considered. These also change the system transfer function in ways which must be taken into account by the feedback management system. Thus, feedback management requires a holistic approach: for optimum performance, the feedback canceller has to be integrated and tuned very carefully to the rest of the adaptive control and signal processing systems of modern hearing instruments. Consequences of feedback on sound quality An important issue to consider is the impact of acoustic feedback on the sound quality of the target signal. Commonly, the implicit assumption is that feedback equals whistling similar to a pure tone. However, this is only the case when the feedback signal is well beyond the critical feedback threshold, i.e. over-critical. When the system is still under-critical but approaching the feedback threshold, the frequency characteristics of the hearing instrument begin to change and a marked impact on sound quality occurs: the instruments may sound rough, modulated or harsh, i.e. artifacts and distortions start to occur.

34 Feedback management Different approaches to feedback management have been introduced in hearing instrument technology [Dillon 2001]. The most successful approach to date is the adaptive feedback cancelling system based on feedback phase inversion. Feedback Phase Inverter Today, state-of-the-art microprocessors allow the implementation of powerful signal processing strategies for effective cancellation of acoustic feedback. Most modern feedback cancellers are based on an inverted phase approach. In this approach, sound waves are cancelled out by their own 180 phase inversion. This is the only technology able to remove feedback without gain reduction. The algorithm comprises two steps: Estimation and modelling of the feedback path. Feedback erasure. For feedback path estimation, a high resolution correlation analysis between hearing instrument input and output is performed. The amount of sound leaking from the receiver back to the microphone is indicated by the result of this correlation analysis. To achieve cancellation, a phase inverted signal with the same frequency content as the feedback signal is generated. Due to destructive interference, the feedback signal is efficiently eliminated without gain reduction. Feedback phase inversion or cancellation has become an accepted and proven method of removing feedback. However, feedback management systems should only be active where and when needed. In programs where feedback is less likely, the feedback phase inverter setting should be less aggressive than in programs where there is greater chance of feedback. While effective at suppressing feedback, many algorithms may be subject to artifacts if the feedback path estimation mechanism falsely identifies other sounds as feedback. This is affected by the degree of feedback phase inversion. As some listening situations have a greater chance of feedback than others, traditional feedback cancellers must have an effective means of ensuring the right balance between feedback suppression, sound quality and effective gain applied by the hearing instrument. The major design criterion for a traditional feedback management system is to find the optimum balance between these three performance dimensions. Different optimized settings should be applied in different listening conditions in order to achieve optimal performance of the entire hearing instrument system. For example, feedback suppression might be given greater priority over sound quality when used together with a telephone as the likelihood of feedback occurring in this situation is greater. On the other hand, when the music program is activated, the adaptive feedback phase inverter ensures optimum sound quality by reducing feedback suppression settings as there is less likelihood of feedback occurring is less. Modern feedback management systems may alleviate feedback, but most will incorrectly identify naturally occurring tonal or correlated signal components ( entrainment ) as feedback and subsequently create unpleasant artifacts. This not only has an impact on the sound quality of the hearing instruments but also limits the amount of applicable gain in the system. Existing systems could be parameterized to reduce feedback more effectively and faster if hearing instrument wearers could tolerate more artifacts and thus a much poorer sound quality. In order to overcome the system performance limitations, it is necessary to precisely identify and distinguish feedback from other tonal signal components. WhistleBlock Technology revolutionary feedback identification and elimination WhistleBlock Technology is a significant step forward in feedback phase inversion by enabling feedback cancellation with much higher effectiveness and precision. It profits from a state-of-the-art feedback identification and tagging module. This module is able to instantly differentiate between true feedback and naturally occurring tones, such as music. Figure 1 shows the trade-off between performance of feedback cancellation systems in terms of added stable gain vs. sound quality. With existing feedback cancellers, increased feedback suppression results in decreased sound quality. The feedback canceller applying WhistleBlock Technology eliminates this trade-off. Higher stable gain is achieved with the same sound quality. Accurate identification of sounds that have re-entered the system as true feedback allows for a precise feedback cancellation strategy, blocking feedback without impacting speech clarity or sound quality (see figure 2). By distinguishing feedback components from other correlated tonal components in the signal, it is possible to apply significantly more aggressive feedback cancellation techniques without creating unwanted artifacts. Better sound quality With WhistleBlock Without WhistleBlock Added stable gain Figure 1: Qualitative representation of the trade-off between sound quality and added stable gain in existing feedback cancellers. Thanks to WhistleBlock Technology, significantly more gain can be added without compromising sound quality.

35 Figure 2: WhistleBlock Technology profits from a state-of-the-art feedback identification and tagging module. This technology is able to instantly differentiate between true feedback and naturally occurring pure tones, such as music. Accurate identification of sounds that have re-entered the system as true feedback allows for a precise feedback cancellation strategy which eliminates the correlation between input and output signals, blocking feedback without impacting speech clarity or sound quality. WhistleBlock Technology performance assessment For assessing the performance of feedback management systems, several different aspects and quality dimensions have to be taken into account. Freed and Soli (2006) and Merks et al. (2006) suggest: (i) added stable gain / effective gain, (ii) effective gain applied, (iii) reliability and speed of feedback detection mechanism, (iv) sound quality. The performance of a feedback management system can be assessed by answering the following questions: How effective is the algorithm at preventing feedback? How effective is the algorithm at reducing sub-oscillatory peaks in the frequency response? Does the algorithm sacrifice gain in any frequency bands? How robust is the algorithm when presented with tonal input signals? In order to answer these questions and assess the performance of feedback management systems, it is necessary to use a reproducible and realistic test set-up. This can be done by placing an artificial head in a sound treated box and using a linear motor to move an object reproducibly close to the ear and the hearing instrument (figure 3). This allows for a realistic and reproducible simulation of different feedback conditions. Using this test set-up, the performance of the new WhistleBlock Technology was compared to other commercially available products with respect to: Added stable gain or excess gain. This measure shows how much more gain can be applied with the feedback canceller turned on versus feedback canceller turned off. Sound quality: considers the amount of artifacts occurring. Robustness in terms of distinguishing real feedback sounds from tonal signal components. Added stable gain Figure 4 shows the added stable gain or excess gain measured for five different devices with the test set-up described above. The instruments were equalized to produce the same amount of gain. It is clearly visible, that the new feedback management system provides significantly more stable gain. Especially in the frequency range most susceptible to feedback, between khz, this new technology provides the largest amount of added stable gain. Figure 3: Test set-up for assessing the performance of feedback management systems under reproducible conditions.

36 Added stable gain [db] WhistleBlock Technology Competitor 1 Competitor 2 Competitor 3 Competitor 4 Competitor Frequency [Hz] 1k 1.5k 2k 3k 4k 6k 8k Figure 4: Added stable gain for different hearing instruments. WhistleBlock Technology available in the Exélia and Naída product lines shows by far the greatest or added stable gain especially in the most critical spectral region between 1.5 khz and 3 khz. Sound quality Figure 5 compares the amount of artifacts, i.e. the sound quality, of a hearing instrument equipped with WhistleBlock Technology with a competitive device that uses another modern feedback cancellation scheme. The left graphs show the results of the device with WhistleBlock Technology. Two sets of spectrograms of the amplified output signals were measured: the first with no object close to the ear and the second one where a solid object was held WhistleBlock Technology about 2 cm away from the pinna. The top row shows the results with the feedback management systems turned off, the bottom row shows the results with feedback management turned on. The red areas in the graphs indicate the occurrence of feedback components or artifacts. With WhistleBlock Technology, hardly any artifacts occur while for a popular competitive system significant artifacts are still present. Competitive Product Feedback Canceller off f [Hz] Feedback Canceller off 6000 Feedback Canceller on f [Hz] Feedback Canceller on Red = feedback Blue = no feedback t[s] Figure 5: Comparison of the presence of artifacts, i.e. the sound quality, of two different feedback suppression systems: left graphs are for WhistleBlock, right graphs for acompetitivesystem. Top row: feedback management turned off. Bottom row: feedback management turned on.

37 Entrainment The last parameter to discuss is entrainment. This phenomenon creates unpleasant artifacts when the feedback cancellation system falsely identifies highly correlated tones as feedback and generates a phase inverted signal. Figure 6 compares the artifacts created due to entrainment of a device with and a device without WhistleBlock Technology. A ring tone was played through a hearing instrument with the respective feedback cancellers turned on. The competitor s feedback canceller (lower panel) produces modulation artifacts, which results in sidebands around the spectral peaks. When using WhistleBlock Technology (upper panel) no entrainment artifacts occur. WhistleBlock Technology can correctly identify tonal input signals and does not take any disturbing counter actions. WhistleBlock Technology Competitive Product Figure 6: Spectrogram showing the amount of artifacts in products with and without WhistleBlock Technology. A ring tone is played and the output of the hearing instrument recorded. The top graph shows the measurement results of a device with WhistleBlock Technology. The yellow and red areas indicate the presence of artifacts that lead to poor sound quality. It is clearly visible, that WhistleBlock Technology leads to significantly less artifacts due to entrainment.

38 Summary WhistleBlock Technology, now available in Exélia and Naída products, will yield optimal performance for many different hearing instrument families and styles. WhistleBlock Technology achieves significant improvements for added stable gain, improved sound quality and reduced entrainment effects when compared to other competitive schemes of feedback cancellation. WhistleBlock Technology achieves unprecedented amelioration of one of the major complaints of hearing instrument users, effectively eliminating feedback without introducing annoying artifacts.

39 Phonak Insight WaterResistant Versatility Introduction There has long been a wish to have hearing instruments that are resistant to water, sweat and dirt. Behind this wish lies the need to be able to wear hearing instruments in all situations without having to worry. This means not having to think about whether you can wear them in heavy rain, during sports or while swimming. Even if accidentally worn in the shower, there is peace of mind and security in knowingthat the sensitive electronics of the hearing instruments are well protected. In a nutshell, there is the simple desire for a carefree life, hearing without limits in all situations. Manufacturers have long pursued the development of hearing instruments that are resistant to water, sweat and dirt. In 2000, Phonak introduced the first product with special elastomer seals, which gave greater freedom and reliability, standing out from the crowd of hearing aids produced by other manufacturers. The next big step forward was the introduction of Naída in the first water resistant Phonak hearing instrument. And now, in 2011, the Phonak Spice+ H2O hearing instruments set a new standard in water resistance and durability they have been awarded a rating of IP67 on the IEC60529/ EN60529 ingress protection standard. IP Certification IP certification has long been used in the household appliance industry. IP stands for "Ingress Protection", i.e. "resistance to penetration" and rates the suitability of appliances under different environmental conditions. In order to be able to compare different products with one another, a range of standards was introduced for various industries. No uniform standard has been introduced for small electronics such as hearing instruments. Consequently, the hearing aid industry has adopted the standard IEC60529/EN60529 in order to demonstrate how water resistant hearing instruments are. This standard is designed to provide protection by enclosures and is applied in the case of appliances such as washing machines. Certification is performed by independent institutes that test and evaluate the products. The result indicates the state of the device after it was exposed to each test condition. The numbers in the IP code indicate the degree of protection the device achieved. The first number in the code indicates the level of protection against solid objects, such as dust, while the second number indicates the level of protection against liquids or moisture. Table 1 lists the numbers and explains what they mean. Table 1: Numbers of the IP certification

40 The new Spice+ H2O hearing instruments: Phonak Ambra M H2O, Solana M H2O, Cassia M H2O, Naída S CRT and Nios S H2O all achieved the IP67 rating. IP67 indicates that the hearing instrument was not damaged beyond repair after 8 hours in a dust chamber, or after being immersed in 1 meter (3 feet) of water for 30 minutes (as defined by IEC60529). Figure 1: Dust chamber Figure 2: Water tank Legend for Figures 3a & 3b 1) Housing top shell 2) Acoustic coupling 3) Housing base 4) Protective membrane - battery compartment 5) Seals 6) Protective membrane - microphone 7) Microphone inlet with protective membrane Figures 3a & 3b show the components of the housing. The Spice+ H2O housing is made up of two parts: the top shell (1) and the base (3). When both parts are put together, a double wall is formed which provides excellent stability. The double wall also averts the ingress of moisture, thus preventing any water from reaching the sensitive electronic parts of the device. Figures 1 & 2: Graphic representation of the IP Certification test setup This certification is a good indication of the dust and water tightness of a device. While IP67 accords these hearing instruments a high degree of protection, it must be kept in mind that these tests were not specifically designed for hearing instruments, and the way they were tested does not necessarily reflect real life situations. Such real life situations are discussed later in the Everyday use section. WaterResistant Spice+ H2O hearing instruments The demands on the new hearing instrument housings were very high, as the devices have to be extremely resistant to dust, dirt, sweat, water and moisture. These demands have been successfully met through the combination of three elements: The housing design, the materials used and nano-coating. Figure 3a) Figure 3b) One effective way to obtain a completely sealed case is with injection molding processes used to produce the housing, so there are no edges or transitions. However, this is quite difficult to achieve in with micro hearing instrument housings. So joints on the housing must be kept to a minimum and properly sealed. The few edges of the Spice H2O products have an additional seal. Using two-component injection molding, the soft sealing material applied to the seams of the housing and the solid material of the housing are fused together, forming a single unit from the two layers. The material makes the critical edges very tight, thus protecting against the entry of any foreign substances. The seal is clearly visible within the battery housing (5), but is also used at the edges of the housing top shell. To ensure that the top shell and base stay together, they are fixed with pins. What is unique about this is that the pins neither penetrate the silicon layer nor come into contact with the interior of the device. This prevents water from entering the housing via the pins. Three other parts of a hearing instrument also require a specific solution: acoustic coupling (tone hook, Slim Tube or external receivers) microphone protection battery compartment The connection between the housing and the acoustic coupling (2) remains a screw connection, in the case of both the Phonak M H2O and Nios S H2O devices. In contrast to previous housings, the threads are embedded in the frame. This makes it possible to install additional sealing rings for a tightlysealed surface. The Naída S CRT has a proven plug-in connector for the acoustic coupling of the external receiver. A special locking mechanism encloses the wire of the external receiver and prevents the ingress of water, sweat and dust. Figure 3: a) Spice+ H2O housing b) Explosion view of the Spice+ H2O housing September /4

41 One of the biggest challenges is ensuring microphone protection, as the microphone inlets must remain open to maintain the acoustic properties of the microphones. In the Spice+ H2O devices, this problem was solved by means of a triple-layer microphone cover. The first layer is part of the housing top shell and consists of two adjacent inlets (7), which are visible from the outside. The top shell with the two inlets is nano-coated. This combination of narrow inlets and nanocoating protects the microphones below from larger dust and dirt particles to which the hearing instrument is exposed in the air or through direct contact. This first layer also prevents scratching noises caused by fingernails or contact with hair. The second layer consists of a special membrane that fits into the inside of the housing top shell directly below the inlets. The membrane is acoustically transparent and air-permeable. It protects against moisture and smaller dirt and dust particles. It consists of a tightly woven fabric which is both hydrophobic and lipophobic; i.e. it is both water- and grease-resistant. These properties make the membrane an effective protection against water and foreign bodies. The first and second layers are part of the top shell and, if necessary, can easily be replaced by hearing care professionals. The same protective fabric forms the third layer on the base of the housing above the microphone inlets (6). This ensures that no particles or water get into the microphone inlets. This layer cannot be replaced by a hearing care professional. This innovation in the field of microphone protection is acoustically stable. The battery compartment of a hearing aid also presents challenges. Nowadays, mostly zinc/air batteries are used in hearing instruments. They require a constant air supply in order to work. The battery compartment has a vent to ensure continuous air supply. At the same time, however, the vent increases the risk of the ingress of water. To prevent this, it is also covered with an air-permeable, water- and dirt-repellent membrane (4), with an even smaller pore size than the membrane for microphone protection. Together with its hydrophobic and lipophobic properties, this provides excellent protection. The battery compartment must be opened when the battery is changed, which represents a further risk factor. A seal (5) is incorporated using two-component injection molding and tightly seals the battery compartment. The sturdy locking mechanism of the battery compartment also ensures a tight seal. In the development of a water resistant housing, the use of FM technology was also taken into consideration, as these hearing instruments are particularly suitable for children and power users. The ML15i FM receiver is integrated in the design and fits all Spice+ H2O devices and is also water resistant. The AS15 universal FM solution with audio shoe and universal FM receiver are not water resistant. Everyday use IP67 confirms that the hearing instrument is highly resistant to dust and water. But what does this actually mean for everyday life? Can an architect wear his hearing instruments, communicating at the dusty building site without needing to worry? Can hearing instruments withstand an unexpected rain shower? We will now look at these questions in more detail using selected examples. The first digit of the IP certification indicates the level of protection against foreign objects. Common examples for this application are working environments such as construction sites, workshops or a farm. In these environments, hearing instruments are often exposed to dust and dirt of different particle sizes. Hearing instruments are also exposed to dust and dirt when working in the garden. As a precaution, many people tend to not use them when gardening. Spice+ H2O hearing instruments can be worn without worry in such dusty environments. All housing parts of the Spice+ H2O devices are nano-coated. This forms an additional repellant layer, so water and dirt just pearl off. Thanks to the combination of outstanding design, sealing elements and protective materials, the Spice+ H2O hearing instruments provide a new benchmark in dust and water protection. September /4

42 The second digit indicates the level of protection against water and moisture, as occurs, for example, in kitchens or bathrooms. Outdoor activities, such as taking the dog for a walk or jogging can often expose the devices to moisture, rain or sweat. Anyone who enjoys being active or doing sports will appreciate these devices, as they offer very high level of moisture protection. People typically do not shower while wearing their hearing instruments, as the sound of the water can be quite loud. However, there is no cause for concern, if someone takes a shower after doing sports and forgets to remove the hearing instruments. The sensitive electronics of the device are well protected. Even occasional submersion in water, as happens with children playing in the bath is no problem. Parents can play and splash, while still communicating with their child. Contact For more information, please contact: Gavin.Buddis@phonak.com Jana.Schimmelpfennig@phonak.com In fact, the devices are ideal for children who want to play, without their parents worrying about their hearing instruments. Children want to explore their surroundings and experience the fascination of water fountains, pools, puddles, as well as sandpits and playgrounds. Children do not want to miss out on anything at day care or on school outings. The Nios S H2O, in combination with the integrated FM receiver, provides care-free protection and security. To maintain this level of protection, does require consistent care and maintenance. The more a device is used in wet and dusty environments, the higher the degree of maintenance that is required. It is recommended to gently rinse the instrument with fresh water after contact with dirt or moisture and then dry the outside and the battery compartment with a soft cloth. There are also limits. Strong forces from contact with water such as during waterskiing or surfing, can damage the microphones. The hearing instruments are not suitable for diving, due to the rapidly increasing water pressure. The pressure damages the microphones, in particular, significantly impairing the functioning of the device. And, with the zinc/air battery being deprived of oxygen under water, the batteries soon fail. Fortunately, they can be easily replaced. Validation subjects confirmed the consistent functioning of the device through exposure to water. But contact with soap should be avoided where possible. Soap can affect the hydrophobic and lipophobic properties of the membrane and compromise the protection. In addition, soap blocks the air circulation of the battery compartment. Although a high standard was achieved with IP67, Spice+ H2O instruments are referred to as waterresistant, not waterproof. They offer hearing impaired people greater freedom. Wearers can confidently embrace all situations, communicating without restrictions. September /4 Phonak AG, All rights reserved

43 Notes:

44 Life is on We are sensitive to the needs of everyone who depends on our knowledge, ideas and care. And by creatively challenging the limits of technology, we develop innovations that help people hear, understand and experience more of life s rich soundscapes. Interact freely. Communicate with confidence. Live without limit. Life is on