Ce6ebe523a57e5995075d202c60ff8cdfc5d

1

On the analysis and improvement of hearing aid devices Diogo Rafael Amado Almeida

Abstract— Hearing Aid devices have come a long way since the implementation of DSPs, but their users still feel the disadvantages of having to use one on a daily basis. This devices fail to simulate the complex ability of the human hearing system, mostly because there is an incredible amount of acoustics involved, and also, sound cannot be reproduced exactly how it is by digital systems. Among the several problems reported by the users the most common are feedback effects, occlusion effects and sense of disorientation and discomfort in loud environments due to the amplification of all the sound around the user. This research focus on directional sound capturing which will be highly valuable for loud environments, since this is the biggest complain of hearing aid users. The ultimate goal is that in a noisy situation the user only listens to his point of interest, and everything else is filtered out. I. INTRODUCTION

E

nhancing Hearing Aids is an area with a lot of different approaches, this Project focus on solving the most common complain of Hearing Aid users, loud environments. What happens is that the natural hearing system filters the information relevant to the individual, so, if a person is having a conversation with another person in a room, and around them is a lot of noise that might include for example, other conversations or music, the individual is still able to stay focused only on the information that he pretends, the conversation. This situation is not the same for Hearing Aid users since the device does not know what information it is supposed to filter, so what happens is that all the sound is amplified equally, thus making it very uncomfortable for the user to be in this situation. The reason that the natural hearing system cannot do the same type of filtering of information after the sound is processed by a Hearing Aid device is because naturally the sound is presented in a three dimensional form, where with a Hearing Aid or even two the sound comes from one or two static sources, therefore the sound seems to come all from the same position. So the natural conclusion is that the Hearing Aid must be responsible for the filtering of the information that the user pretends, and this is possible with directional sound capturing. Most of the times the user is facing the direction of the sound intended to be listened, so the angle of interest is acquired,

regarding the distance of interest, this can be achieved by several methods presented bellow. After this sound filtering, the user will be able to have a conversation or listening to a song as good or possibly even better than an individual with a perfect hearing capability would in a loud environment, since now, all the irrelevant sound is not being amplified.

II. STATE-OF-THE-ART In the market there are around ten different main types of Hearing Aids, the main difference between those types is reflected in the way the acoustics work and the size of the device. The size is important because the bigger it is, the most processing capacity can be implemented, but the portability decreases and the comfort for the user is reduced. The shape of the earpiece is going to affect acoustics, so there are different approaches depending on the needs of the user. An example of how the shape can affect acoustic is the occlusion effect, this occurs when the earpiece blocks the ear channel making it hard for air to flow freely, so the sound generated by our body, for example chewing, is going to reverberate much more than it should. The biggest type of Hearing Aids are “body worn aids”, not as used nowadays due to the existence of much smaller devices, but they have the potential for much greater battery life and better processing capabilities, but they wouldn’t be good for children or individuals with a very active life style. One of the most famous Hearing aids are called “behind the ear”, they have medium size and are supported by the ear it self, they are practical to remove and operate, but are visible to the naked eye. Esthetics is also a concern, so there are much smaller types of Hearing Aids like “In the ear”, “Invisible in canal hearing aids” and “Receiver in the Canal”. Regarding the shapes, there are Hearing Aids specifically designed to improve the acoustics, “Open-fit devices” are an example, where the earpiece is not blocked thus reducing the occlusion effect discussed previously. Due to their small size “Invisible in canal hearing aids” also reduce this effect a lot. Regarding the sound treatment there are several approaches used in Hearing Aids, “Digital Feedback Reduction” which solves a very common problem in the past, feedback, this usually occurred when the volume of the device was to loud or if the user was close to the source of the sound. “Digital Noise Reduction” a process that analyses the sound and reduces the noise, this process is done by averages of the input

2 information or by canceling certain frequencies, and Digital Signal Processing in general and as needed, for example, some users just need the higher frequencies to be amplified or even moved to lower frequencies so they can ear them. Directional Microphones are also implemented in most of Hearing Aids nowadays with the same goal as this Project, enhancing Signal to Noise Ratio, but this is, as stated previously, still the biggest complain amongst the users.

III. CONCEPTUAL DESIGN The main goal is to create a system that the user can activate or deactivate whenever necessary, this will allow for the Hearing Aid to operate normally and when confronted with noise situations, the system is turned on. It is not intended to be an automatic switch to avoid false-positive situations, and this way the user can have full control of the situation. The design is divided in two main areas, angle of reception and depth.

Figure2-Microphones polarities Due to the use of a low angle capture microphone it will be easier to compute the information to reject since the sound of the back of the user is already attenuated by the microphone, so it´s a question of filtering amplitudes. The main part of processing will be calculating the exact center of the two microphones, this way the system can know exactly where the user is facing.

D(n)= Common source of information that we want to keep, center. VL(n),VR(n)= Information that we want to filter out coming from left and right. The depth of interest is not a constant, opposed to the position of the user’s head in relation to the microphones, so it will be user controlled, using the same remote that turns the entire system on/off. Off course this will have limitations due to the sensitivity of the microphones, being the intent minor adjustments in the normal range of a conversation.

Figure1-System overview The angle of reception will be calculated with the use of two physical microphones, one in each ear, and the use of FPGA, several tests will be performed with different types of microphones, but the ones most likely to be used are cardioid due to the low coverage angle, high angle of rejection, high rear amplitude rejection and low ambient sound sensitivity.

Figure3-Remote Sketch

3 The system design will be composed of two microphones and a remote connected to a FPGA that connects to a single speaker. This system operates alone, since it is meant for special situations, but it is possible to adapt to any type of existing hearing aid device where we can use the built in speaker as an output.

VII. CONCLUSION The system will be highly beneficial if it turns out a success, since it is not a passive feature and can be adaptable to any type of Hearing Aid it has no compromise in terms of sound quality. It is a great challenge to solve the most common problem in Hearing Aid devices, but the outcome can be positive to the point of actually changing the life quality of a lot of individuals. REFERENCES [1] Bernarding, C.; Strauss, D.J.; Latzel, M.; Hannemann, R.; Chalupper, J.; Corona-Strauss, F.I. “Simulations of hearing loss and hearing aid: Effects on electrophysiological correlates of listening effort” in Engineering in Medicine and Biology Society,EMBC, 2011 Annual International Conference of the IEEE , 2011 , Page(s): 2319 – 2322 [2] Spriet, A.; Rombouts, G.; Moonen, M.; Wouters, J. “Combined Feedback and Noise Suppression in Hearing Aids” in IEEE Transactions on Audio, Speech, and Language Processing, 2007 , Page(s): 1777 – 1790, Vol.15 [3] Lindemann, E. “Two microphone nonlinear frequency domain beamformer for hearing aid noise reduction” in IEEE ASSP Workshop on Applications of Signal Processing to Audio and Acoustics, 1995 , Page(s): 24 – 27

Figure4-System Sketch

IV. IMPLEMENTATION After the conclusion of the system, this will be implemented on an existing Hearing Aid, this way it will be possible to keep all the existing complex treatment of the sound required by the user on a normal situation and when the user requests the activity of the system the DSP on the Hearing Aid will be turned off and the FPGA of the system will be turned on, using only the speaker of the Hearing Aid, this will return to normal operation after the system is turned off. V. SCHEDULE First Prototype, February, 15,2013 Final Prototype, February, 28,2013 Paper sent for approval, March, 31,2013 Final Paper submitted, April, 30,2013 VI. ANALYSIS The system prototype will be tested in a room with more than 3 sources of sound, all with the same amplitude but different distances and positions. The goal will be to isolate on real time each one of them as a user walks around the room and face each source separately. The sound of the speaker of the system will be recorded so success rate can be analyzed after the experiment. The success rate will be based on the amount of ambient sound present, the higher the worst.

[4] Chazan, D.; Medan, Y.; Shvadron, U. “Noise cancellation for hearing aids” in IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP, 1986 , Page(s): 977 - 980 [5] Puder, H. “Hearing aids: an overview of the state-of-theart, challenges, and future trends of an interesting audio signal processing application” in Proceedings of 6th International Symposium on Image and Signal Processing and Analysis, 2009, Page(s): 1 - 6 [6] Fukane, A.R.; Sahare, S.L. “Enhancement of Noisy Speech Signals for Hearing Aids” in International Conference on Communication Systems and Network Technologies, 2011 , Page(s): 490 – 494 [7] Whitmal, N.A.; Rutledge, J.C. “Noise reduction in hearing aids: a case for wavelet-based methods” in Engineering in Medicine and Biology Society, 1998. Proceedings of the 20th Annual International Conference of the IEEE , 1998 , Page(s): 1130 - 1135 vol.3 [8] Min Li; McAllister, H.G.; Black, N.D.; De Perez, T.A. “Perceptual time-frequency subtraction algorithm for noise reduction in hearing aids” in IEEE Transactions on Biomedical Engineering, 2001 , Page(s): 979 – 988 vol.48 [9] Ellaham, N.; Giguere, C.; Gueaieb, W. “Evaluation of two speech and noise estimation methods for the assessment of nonlinear hearing aids” in 2011 IEEE International Conference on Acoustics, Speech and Signal Processing

4 (ICASSP), 2011 , Page(s): 293 – 296 [10] Wen-Chih Wu; Cheng-Hsun Hsieh; Hsin-Chieh Huang; Chen, O.T.-C. “Hearing aid system with 3D sound localization” in TENCON 2007 - 2007 IEEE Region 10 Conference, 2007 , Page(s): 1 – 4 [11] Yang Min “Design of portable hearing aid based on FPGA” in 4th IEEE Conference on Industrial Electronics and Applications, 2009 , Page(s): 1895 – 1898 [12] Dong-Ook Chung; Won Doh; Dae-Hee Youn; Jae-Yeon Choi; Hyo-Chang Woo; Dong-Wook Kim; Won-Ki Kim “Hearing impairment simulation for the performance evaluation of hearing aid system” in Engineering in Medicine and Biology Society, 1996. Bridging Disciplines for Biomedicine. Proceedings of the 18th Annual International Conference of the IEEE, 1996 , Page(s): 415 - 416 vol.1 [13] Pandey, A.; Mathews, V.J. “Improving adaptive feedback cancellation in digital hearing aids through offending frequency suppression” in 2010 IEEE International Conference on Acoustics Speech and Signal Processing (ICASSP), 2010 , Page(s): 173 – 176 [14] Kamkar-Parsi, A.H.; Bouchard, M.” Instantaneous Binaural Target PSD Estimation for Hearing Aid Noise Reduction in Complex Acoustic Environments” in IEEE Transactions on Instrumentation and Measurement, Issue: 4 , 2011 , Page(s): 1141 – 1154 Vol.60 [15] Young-cheol Park; In-young Kim; Sang-min Lee “An efficient adaptive feedback cancellation for hearing aids” in Engineering in Medicine and Biology Society, 2003. Proceedings of the 25th Annual International Conference of the IEEE, 2003, Page(s): 1647 - 1650 Vol.2 [16] Vicen-Bueno, R.; Gil-Pita, R.; Utrilla-Manso, M.; Alvarez-Perez, L. “A hearing aid simulator to test adaptive signal processing algorithms” in IEEE International Symposium on Intelligent Signal Processing, 2007 , Page(s): 1 –6 [17] Sang Min Lee; Jong Ho Won; See Youn Kwon; Youngcheol Park; In Young Kim; Sun I Kim “New idea of hearing aid algorithm to enhance speech discrimination in a noisy environment and its experimental results” in Engineering in Medicine and Biology Society, 2004. IEMBS '04. 26th Annual International Conference of the IEEE, 2004, Page(s): 976 – 978 Vol.1 [18] Edwards, B.W. “Signal processing techniques for a DSP hearing aid” in IEEE International Symposium on Circuits and Systems, Page(s): 586 - 589 vol.6 [19] Bernarding, C.; Strauss, D.J.; Latzel, M.; Hannemann, R.; Chalupper, J.; Corona-Strauss, F.I. “Simulations of hearing loss and hearing aid: Effects on electrophysiological correlates of listening effort” in Engineering in Medicine and Biology Society,EMBC, 2011 Annual International Conference of the

IEEE , 2011 , Page(s): 2319 – 2322 [20] Pandey, A.; Mathews, V.J. “Offending frequency suppression with a reset algorithm to improve feedback cancellation in digital hearing aids” in IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2011 , Page(s): 301 - 304 (temp) [21] http://www.acesandeighths.com/microphone_evo.html