Harry Levitt- harrylevitt@earthlink.net
Advanced Hearing Concepts
Bodega Bay, CA 94923
Popular version of paper 3aEA1
Presented Tuesday morning, November 30, 2006
4th ASA/ASJ Joint Meeting, Honolulu, HI
The journey leading to the development of digital hearing aids began at
Bell Laboratories in the mid 1960s when a digital computer was used to simulate an experimental high gain telephone with frequency shaping for people with hearing loss. The simulation was off-line at about 100 times real time. About 20 years later, a digital master hearing aid operating in real time was developed using a high speed array processor. The development of wearable digital hearing aids followed soon after as high speed digital signal processing (DSP) chips were developed. The first wearable digital hearing aid was a body worn instrument and was not a commercial success. At this stage, digitally controlled analog hearing aids could be made small enough to be worn on the ear and these instruments were widely used by the end of the decade. During the 1990s, low power DSP chips were developed that were small enough to allow for all-digital ear-worn hearing aids to be developed. The challenge today is to tap the tremendous potential of digital signal processing techniques within the power and size constraints of a practical, wearable instrument.
Preliminaries to the Development of a Digital Hearing Aid
The first electronic hearing aids were large instruments that were inconvenient to use. A major step forward followed the development of miniature electron tubes which allowed for the development of wearable hearing aids. These were body worn units with a wire link to an earpiece containing a hearing aid receiver (loudspeaker). The next breakthrough followed the invention of the transistor. Hearing aids small enough to fit behind the ear were developed and with the continued miniaturization of solid state electronics, hearing aids small enough to fit in the ear canal were developed. Up until now, hearing aids were viewed as no more than miniature personal amplifiers for use by people with hearing loss.
The introduction of digital technology in hearing aids has brought about substantial changes not only in the design of hearing aids, but also in the basic function of a hearing aid. Hearing aids are no longer devices for simply amplifying sound. The first steps towards this remarkable transformation began with the scientific use of digital computers. In the early 1960s researchers at the Bell Telephone Laboratories developed methods for analyzing and processing speech and other audio signals on a large mainframe computer. An important outcome of this research was the development of convenient methods of simulating complex speech transmission systems, such as the voice coders (vocoders) for use on the transatlantic cable. One of the researchers at the Bell Telephone Laboratories reversed the process and used this extremely sophisticated system for simulating a simple amplifier as part of an experiment involving a telephone with high gain for people with hearing loss. It was much like using was a steam roller for pressing a shirt. On leaving the laboratories, the researcher took the basic idea of simulating hearing aids on a computer to a university where strange ideas are known to blossom.
One of the problems with simulating hearing aids on a digital computer at that time was that computers were relatively slow. The time taken to process an audio speech signal was many times longer than the duration of the signal. The large main frame computer used for these simulations also occupied several racks of equipment. It was inconceivable at that time that digital processing of sound would one day be possible in a microcomputer small enough to be worn on the ear. The research done at that time was nevertheless valuable in developing methods of digital signal processing that would one day be used in developing practical digital hearing aids. The most valuable aspect of this work was that it provided new insights and engendered new ways of thinking about how to process sounds for people with hearing impairment.
The introduction of minicomputers in the mid to late 1960s opened the door to laboratory studies of real-time signal processing for people with hearing loss. These machines were not fast enough to process audio signals digitally in real-time, but they were used to control conventional analog equipment for processing of audio signals in real-time. Computer controlled analog systems were widely in the 1970s for studying new forms of amplification for people with hearing loss. Edgar Villchur who had invented the AR bookshelf loudspeaker introduced the idea of multichannel amplitude compression for hearing aids. In this form of amplification, the incoming audio signal is subdivided into a series of contiguous frequency bands and the gain in each band is adjusted such that intense sounds within each band receive less amplification so that they are not too loud while weak sounds receive more amplification so as to be audible above the elevated hearing threshold of the impaired ear. This configuration, a bank of analog filters controlled by a digital unit, was the subject of much research in the 1970s. It would later become the basic architecture of the first hearing aids to employ digital technology.
In 1975, Daniel Graupe reported the development of a six-channel hearing aid with digital control of the gain in each channel. This was a hybrid instrument rather than an all-digital hearing aid, but it was the forerunner of instruments of this type. Graupe went on to develop the Zeta Noise Blocker, a self-adapting filter that automatically attenuated frequency channels containing high noise levels. In 1977 Mangold and Leijon in Sweden reported on the use of a programmable filter in acoustic amplification and subsequently developed a programmable multichannel compression hearing aid using digital control of analog components. This hearing aid also had multiple memories so that, at the touch of a button, the electroacoustic characteristics of the hearing aid could be altered to provide appropriate amplification for a different acoustic environment (e.g., a noisy as opposed to a quiet room).
Youll Need a Friend With a Wheelbarrow.
The 1970s also saw the development of high speed digital array processors suitable for use with minicomputers. Array processors were fast enough to process audio signals digitally in real time and in 1982 an array processor digital hearing aid was developed at the City University of New York. As shown in the photograph, the equipment was relatively large. The box-like units on the right consist of a minicomputer and a digital array processor. The small units on top of the computer are an FM radio transmitter and receiver, respectively. These units provide a radio link with a body worn transmitter and receiver which, in turn, are linked by wire to an ear worn microphone and hearing aid receiver (loudspeaker). As one wag put it, It may be a good hearing aid, but youll need a friend with a wheelbarrow behind you to carry the instrument.
The array processor digital hearing aid was designed as a research tool for exploring the potential of digital signal processing in hearing aids. An important feature of its design was that it could be used to simulate experimental hearing aids. One of the problems in hearing aid research at that time was that any new idea for improving signal processing in a hearing aid required that an instrument be constructed in order to evaluate the technique. This was a time consuming, costly approach. Further, if the prototype instrument did not perform as well as predicted by theory, it was often uncertain whether the theory was in error, or whether hardware imperfections in the prototype were to blame. In contrast, it was much less difficult to simulate different hearing aids and to evaluate them experimentally with the array processor digital hearing aid prior to taking on the more difficult task of actually building a prototype. The array processor digital hearing aid served its design purpose well and was used as a research tool for over a decade investigating new methods of signal processing for use in hearing aids.
The early 1980s saw the development of digital chips dedicated to high speed digital signal processing. These ‘DSP’ chips were fast enough to process speech and other audio signals in real time, but were too large and consumed too much power to be used in wearable hearing aids. It was only a matter of time before their size and power consumption could be reduced to levels that would be practical for a digital hearing aid. Engebretson, Morley and Popelka at the Central Institute for the Deaf (CID) began working on a digital hearing aid in the early 1980s and were the first to develop a practical, wearable digital hearing aid including fabrication of the DSP chips. The CID group later joined with the 3M Company in a consortium to develop digital hearing aids.
In 1987, the Nicolet Corporation introduced the first commercial digital hearing aid. Their first hearing aid consisted of a body worn processor with a hardwire connection to ear-mounted transducers. A behind-the-ear (BTE) digital hearing aid was introduced two years later shortly before the Nicolet Corporation withdrew from the hearing aid market. Although the Nicolet hearing aid was not a commercial success, it demonstrated the feasibility of a wearable digital hearing aid and the race was on for other hearing aid companies to develop commercially viable digital hearing aids. Bell Laboratories developed a hybrid digital/analog hearing aid in which digital circuits controlled a two-channel compression amplifier. Although field trials showed the instrument to be superior to state-of-the-art hearing aids, AT&T (the parent company of Bell Laboratories), for reasons unrelated to the product, withdrew from the hearing aid industry. AT&T assigned the rights to their hearing aid to the Resound Corporation in 1987 and after further refinement the hearing aid was marketed and was an immediate success.
The first commercially available programmable hearing aid, the Audiotone System 2000, was introduced by Dahlberg in 1988. Other hearing aid companies followed soon after with programmable instruments such as the Bernafon PHOX and the Siemen’s Triton. In 1989, the 3M Corporation introduced the Memory Mate based on the design of Mangold and Leijon noted earlier. Within a few years virtually every hearing aid company was marketing one or more programmable hearing aids. These instruments used analog components (amplifiers, filters, limiters) controlled by a digital unit. Digitally controlled analog hearing aids provided some of the benefits of digital hearing aids, such as memory for storing parameter settings, the capability for paired-comparison testing, and convenient selection of appropriate parameter settings for different acoustic environments.
This was a period of rapid change in the hearing aid industry. New hearing aids embodying novel designs with increasingly more complex signal processing were introduced in quick succession by different manufacturers. At the same time, work progressed intensely on developing the ultimate prize – a true digital hearing aid. The main problem in developing a commercially viable digital hearing aid was that of reducing the size and power consumption of the digital chips. Analog chips for hearing aid applications had been refined over many years so that even when digital chips were developed with an acceptable level of power consumption and small enough for use in a practical hearing aid, there remained the problem of competing with analog technology that was already well established with chips of even smaller size and lower power consumption. Advances in digital chip technology continued steadily over the years so that it was essentially a matter of time before digital chips became competitive with analog chips for hearing aid applications.
The turning point came in 1996 when Widex introduced the Senso, the first commercially successful digital hearing aid. The Oticon Company began marketing their digital hearing aid, the Digifocus, immediately afterwards. The DigiFocus was based on the JUMP-1 digital hearing aid platform which had been distributed the year before to audiological research centers worldwide in order to spur independent investigations of how best to implement digital technology in acoustic amplification.
Advantages of Digital Hearing Aids and New Prospects
The first generation of digital hearing aids used a fixed multichannel architecture. As a consequence, these hearing aids were not very different conceptually from the previous generation of hybrid digital/analog hearing aids. Multichannel amplitude compression was a trusted design and the first digital instruments essentially refined this format. As a consequence, early experimental comparisons between digital and analog hearing aids showed only small improvements in performance for the digital instruments. The recent introduction of digital hearing aids with an open architecture has allowed for the development of innovative new designs that take greater advantage of the unique capabilities of digital processing.
Digital hearing aids are much more effective in reducing unstable acoustic feedback (manifested as a loud whistling of the hearing aid) than analog hearing aids. Digital feedback cancellation involves more than avoidance of loud unpleasant whistling, it also reduces major constraints on the overall gain of a hearing aid thereby allowing for greater use of open, more comfortable ear molds and more accurate control of the overall frequency-gain characteristic.
Difficulty understanding speech in noise is the most common complaint of hearing aid users and considerable effort has gone into the development of digital hearing aids with noise reducing capabilities. The most successful approach has been to develop hearing aids with adjustable directional characteristics. Hearing aids have also been developed recently which recognize the direction of an interfering noise so as to adjust the directional characteristics of the hearing aid to attenuate the interference. Automatic sound recognition is not a simple problem and this approach to noise reduction is still in its infancy, but it holds much promise.
An intriguing aspect of automatic sound recognition is to recognize important information bearing components of the speech signal and to amplify these components so as to make them easier to recognize by people with hearing loss. This is not a new idea, but digital signal processing provides the means for implementing these techniques in a practical hearing aid.
The results obtained in field evaluations using a wearable prototype digital hearing aid were found to be substantially different from those obtained in the clinic. These differences were far greater than initially expected and led to a reconsideration of how hearing aids should be evaluated. It was clear that data obtained using traditional methods of hearing-aid evaluation in a clinic are not predictive of real world performance. There are important individual differences in lifestyle that affect hearing aid usage and the conditions under which the hearing aid is used. A promising approach to this problem is the recent introduction of Artificial Intelligence (AI) processing in a hearing aid. One such application is the monitoring of patterns of use, such as volume control settings, daily usage, and types of sounds being received (speech, noise) so as to find the most appropriate settings of the hearing aid for a given user as well as improving counseling.
Despite the progress in developing advanced methods of digital signal processing to reduce the effects of background noise, the use of a remote microphone located close to the speaker’s mouth is still the most effective way of dealing with background noise and room reverberation. This approach, however, requires the use of a hard wire or wireless link between the remote microphone and the hearing aid which is not always convenient. The impact of digital technology on the use of wireless links in acoustic amplification is only beginning to be felt. The recent introduction of novel methods of linking hearing aids to digital wireless telephones (and to the internet) using the Bluetooth digital wireless protocol holds considerable promise, especially since both data and control signals can be transmitted as well as audio. This will allow for convenient monitoring of the acoustic environment, efficient data logging and remote, automatic adjustment of a hearing aid by a powerful central processor. The landscape for future developments is boundless.
In summary, the development of digital hearing aids began slowly, but increased rapidly as advances in digital technology gained momentum. Each new advance introduced new ways of processing audio signals which not only resulted in improved methods of acoustic amplification but, more importantly, introduced new ways of thinking regarding the fundamentals of acoustic amplification. In the pre-digital era, hearing aids did little more than amplify sound. Today, the hearing aid is a far more complex instrument in which amplification is combined with advanced forms of signal processing for speech enhancement, noise reduction, self-adapting directional inputs, feedback cancellation, data monitoring and acoustic scene analysis, as well as the means for a wireless link with other communication systems.