Specification and calibration of acoustic short-term stimuli for objective audiometry

نویسندگان

  • Thomas Fedtke
  • Johannes Hensel
چکیده

Many applications in objective audiometry measure and evaluate responses of the hearing system to acoustic shortterm stimuli. One of the earliest and best known among these signals of short duration is the 'reference pulse' specified in standard IEC 60645-3. This definition of an electrical reference signal dates back to times when audiometric equipment essentially consisted of analogue components used in laboratory set-ups. Therefore, overshoot caused by the limited bandwidth of the electrical signal was not considered in this definition. Modern equipment, however, often uses audio-frequency DA converters, which imply a limited bandwidth of the signal path. Even if their sample rate is chosen in such a way that the physiological and psychoacoustic effects of the pulse do not differ from those of the quasi-non-band-limited pulse, the transients caused by DA conversion and bandwidth limitation lead to a time response of the pulse conflicting with the current IEC specification. In order to resolve this problem, a modified definition of the reference pulse by detailed time-domain tolerance diagrams is discussed. Meanwhile, in addition to the 'signals of short duration' specified in IEC 60645-3, a variety of short-term stimuli with considerably different temporal characteristics is available for audiometry. ISO 389-9 explicitly allows the application of these signals, provided that they are clearly specified by the manufacturer. The concept of expressing their reference thresholds by means of peak-to-peak equivalent Reference Equivalent Threshold Sound Pressure Levels (peRETSPLs) is easy to implement, even with rather simple instrumentation. However, this concept results, for some stimuli, in calibration values which do not at all correlate with either the behavioural hearing thresholds or the spectral energy of the signals. Attempts at improving the calibration procedure with respect to these 'inconsistencies' are sketched out in this paper. INTRODUCTION Short-term stimuli for objective audiometry, covering the range from basic rectangular reference pulses to sophisticated signals, e.g. chirps optimized to evoke high potentials for hearing screening applications, need to be exactly specified and properly calibrated in order to ensure consistent and comparable measurement results. The basic signals of short duration are specified in standard IEC 60645-3:2007 “Electroacoustics – Audiometric equipment – Part 3: Test signals of short duration” [1]. The definition of the electrical reference pulse for generating click signals is not fully applicable to modern equipment using audio-frequency DA converters, because these imply a limited bandwidth of the signal path, resulting in overshoot and transients. This problem is discussed in the first part of this paper. Reference hearing thresholds for short-term stimuli are usually expressed by peak-to-peak equivalent Reference Equivalent Threshold Sound Pressure Levels (peRETSPLs) according to ISO 389-6. Though this concept is easy to implement, even with rather simple instrumentation, one of the procedure's drawbacks is the lack of coherence with the behavioural hearing threshold. Even signals with very similar power spectra yield widely spaced reference threshold sound pressure levels. The second part of this paper focuses on practical examples and attempts at improving the calibration of short-term stimuli. SPECIFICATION OF ELECTRIC REFERENCE PULSE Standardized reference pulse The specification of the electric reference pulse in IEC 60645-3 dates back to times when audiometric equipment essentially consisted of analogue components. Overshoot caused by the limited bandwidth of the electrical signal was not considered in the specification: "The reference pulse ... shall be an electric rectangular pulse (single monophasic rectangular wave) of (100 ± 10) μs duration with rise and fall times less than 25 μs" [1]. The duration is defined as the time interval between the instantaneous values of 50 % of the signal amplitude; rise and fall time correspond to the time interval between 10 % and 90 % of the signal amplitude, respectively. These requirements are easy to meet if the pulse is generated by means of an electrical function generator with a bandwidth which is much wider than the audio-frequency bandwidth. The oscillogram of a reference pulse generated in this way is shown in Figure 1. The upper waveform (green) is the function generator output signal and the lower one (purple) is the signal measured across the electrical terminals of an audiometric headphone driven by a conventional audiometer. 23-27 August 2010, Sydney, Australia Proceedings of 20th International Congress on Acoustics, ICA 2010 2 ICA 2010 Figure 1. Oscillogram of an electrical reference pulse. Upper waveform: function generator output signal; lower waveform: electrical terminals of audiometric headphone. Figure 1 proves that the requirements described above are readily observed by the signal at the headphone terminals. There is neither noticeable overshoot nor any other transient oscillation. Impact of limited bandwidth Modern audiometric equipment, however, often uses audiofrequency DA converters, which imply a limited bandwidth of the signal path. Usual audio sampling rates give rise to remarkable overshoot and transient oscillations of the electrical output signal. This means that the waveform of the pulse driving the audiometric transducer does not comply with the current IEC 60645-3 specification, which does not allow for any overshoot. Based on a rough estimation of the achievable rise and fall times as 80 % of the sampling interval, the requirements of IEC 60645-3:2007 could just be met using a sampling frequency of 32 kHz. Audiometric equipment which partly uses sampling rates as low as 16 kHz is definitely unable to generate standardized reference pulses. For such devices, reference hearing thresholds determined using reference pulses according to IEC 60645-3:2007 are not applicable. Provided that the pulse is appropriately designed in the frequency domain, aiming at optimizing slew rate and overshoot in the time domain, conventional audio DA converters (sampling rate 44,1 kHz), however, are definitely able to generate pulses compliant with the duration and rise/fall-time specifications of IEC 60645-3. Figure 2. Spectrum sin(x)/x (x = π • f / f0; f0 = 10 kHz) for a 100 μs reference pulse with an upper limiting frequency of 20 kHz. For a series of listening tests using such a conventional audio DA converter, the rectangular pulse was designed in the frequency domain by the spectral representation of an ideally rectangular pulse with a duration of 100 μs: the sinc function sin(x)/x (with x=π • f / f0; f0=10 kHz). The second zero of this function is located at 20 kHz, and all higher frequencies were completely blanked out, see Figure 2. Hence, the pulse has an upper limiting frequency of 20 kHz. The waveform of this pulse signal, measured across the electrical terminals of an audiometric headphone, easily meets the requirements for duration, rise and fall time, as shown in Figure 3. Figure 3. Waveform of the generated reference pulse (44,1 kHz sampling rate); measured by means of an oscilloscope at the headphone terminals; duration: 98 μs; rise and fall time: 22,8 μs; overshoot: 21 %, transient oscillations: 10 % of amplitude. Overshoot and transient oscillations with the observed magnitude do not essentially influence the technical and psychoacoustic effect of the (acoustical) click, for example, the corresponding reference hearing threshold. This was demonstrated by the listening tests [2]. Time-domain tolerance diagram A detailed time-domain tolerance diagram was derived from the band limited reference pulse described above. It adds tolerances for ripple and transient oscillations to the basic duration and slew-rate requirements of IEC 60645-3, see Figure 4. Figure 4. Time-domain tolerance diagram for the electrical reference pulse. This tolerance diagram allows for an acceptable extent of overshoot and transient oscillations, but, at the same time, sets limits which do not unduly affect reference hearing thresholds and other reference values, determined with pulses according to the current specification in IEC 60645-3:2007. 10

برای دانلود رایگان متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید

ثبت نام

اگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید

منابع مشابه

Short-Term Effect of Gabapentin on Subjective Tinnitus in Acoustic Trauma Patients

Introduction:Although several treatment approaches have been proposed for tinnitus, there are currently no Food and Drug Administration (FDA)-approved agents available to treat this condition. In this study, we evaluated the effect of gabapentin on the sensation of subjective tinnitus in patients with acoustic trauma referring to the ear, nose and throat (ENT) clinic of Taleghani Hospital durin...

متن کامل

کاربرد فیلم در القای هیجانهای خوشایند و ناخوشایند و تغییر بازتاب از جا پریدن

  Objective : the studies indicate that the "eye-blink" components of "acoustic startle reflex" can be modulated through emotionally slide stimuli. Pleasant stimulants reduce eye-blink amplitude, whereas unpleasant stimulants enhance them. Method: the present study examines the modulation of the acoustic startle reflexes through a short film clips (2-min), classified as pleasant, unpleasant and...

متن کامل

Brain stem evoked response audiometry of former drug users.

UNLABELLED Illicit drugs are known for their deleterious effects upon the central nervous system and more specifically for how they adversely affect hearing. OBJECTIVE This study aims to analyze and compare the hearing complaints and the results of brainstem evoked response audiometry (BERA) of former drug user support group goers. METHODS This is a cross-sectional non-experimental descript...

متن کامل

Status of sound field audiometry among audiologists in the United States.

Results of a national survey on sound field usage and sound field calibration are presented. Respondents to the survey were 418 ASHA certified audiologists, who answered questions on demographic variables, sound field testing conditions, and sound field calibration procedures. Demographic characteristics of the respondents indicate that they were representative of ASHA certified audiologists. E...

متن کامل

[Temporal resolution: analysis in term and preterm preschoolers].

BACKGROUND auditory processing. AIM to verify the hearing behavior of temporal resolution in children with ages from five to six years, who were born preterm, with no evidence of neurological alterations and to compare this behavior to that observed in children of the same age, who were born at term, with low risk for developmental disorders, taking into consideration the variables of: thresh...

متن کامل

ذخیره در منابع من


  با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید

عنوان ژورنال:

دوره   شماره 

صفحات  -

تاریخ انتشار 2010