This section shares benchmarking and related research information from inside Starkey Hearing Technologies, including detailed methodology used in data collection.
The Hearing In Noise Test (HINT)
Poor speech understanding in noise is one of the most commonly heard complaints that hearing healthcare professionals hear from their patients (Kochkin, 2002). One proven method of addressing this complaint is the use of directional microphones (Ricketts, 2001). As such, the usage of an effective directional microphone is an important design focus in today's hearing aids. For this reason, all new Starkey hearing instruments are evaluated on the amount of benefit that can be obtained for speech recognition in noise. The Hearing in Noise Test (HINT) is one example of a standardized test that evaluates speech recognition in noise (Nilsson et al., 1994). The HINT is routinely administered as a part of Starkey's clinical validation battery.
This section provides a brief overview of the HINT, along with a description of the typical methodology that Starkey utilizes with the HINT to evaluate benefit for speech understanding in noise with directional microphone technology for its new products. The final section provides a specific example of how the HINT was used as part of a clinical research trial to evaluate Starkey's directional microphone technology in a receiver-in-the-canal product.
Overview of the HINT
The Hearing in Noise Test (HINT) is a standardized speech reception test that measures sentence recognition in background noise. The HINT is composed of 250 sentences, which are categorized into 25 lists. An additional three lists are included for practice. The sentences for the HINT are adapted from 336 Bamford-Kowal-Bench (BKB) sentences (Bench & Bamford, 1979), and are revised from the original British versions to equate length and content for American English use.
The HINT is an adaptive speech recognition task. In other words, the level of each sentence is adjusted based on the response of the listener. The speech presentation level is decreased after each correct response, which raises the level of difficulty for the next sentence on the list. Conversely, the presentation level is increased after each incorrect response, which reduces the difficulty for the following sentence. The level of the noise is held constant, thus adapting the presentation level of the sentences varies the signal-to-noise ratio (SNR).
The adaptive nature of the HINT ensures that the listener will approach a 50% correct response rate. Note that the HINT sentences can also be administered in quiet to assess sentence recognition in quiet. In this case, a threshold for sentence recognition is obtained. If the test is administered in noise, adapting the SNR allows for the estimation of a SNR threshold for speech recognition in noise. As the SNR score decreases, the listening conditions become more difficult. For instance, if a hard-of-hearing listener is able to understand speech at an SNR of – 3 dB in an omnidirectional hearing aid setting they may be able to understand speech at an SNR of -6 dB when the microphone setting has been changed to directional. The change in microphone allows the listener to understand speech under more adverse conditions, thus improving their speech recognition ability in noise.
One main benefit of using this type of an adaptive test is that the test results for each patient are obtained at the same performance level of approximately 50%. This allows for comparisons of scores across different patients and different conditions for the same patient. For instance, the SNR score for a particular patient can be compared to listeners with normal hearing. In evaluating benefit with hearing aids, the SNR score for an unaided condition can be compared to a number of aided conditions. A variety of hearing aid settings can be tested to evaluate benefit for different settings such as an omnidirectional microphone setting vs. a directional microphone setting. The adaptive method is also ideal because ceiling and floor effects are less likely to occur. In other words, one does not have to worry about scores at or near 0% and 100%. The only exception may be for patients who have very poor speech discrimination in quiet where a speech threshold is not obtainable even at a very high SNR.
The HINT procedures are documented in many peer reviewed publications as well as the commercially available CD audio package. Specifics on the development and application of the HINT test can be found in Nilsson, et al. (1994). Examples of the use of the HINT in a variety of research applications can be found in a variety of publications as well (Hornsby & Ricketts, 2007; Nordrum et al., 2006; Ricketts et al., 2001).
While test conditions and procedures may vary based on the research goal, the typical methodology for evaluation of speech in noise performance in Starkey clinical trials is described below.
In the case of a clinical trial, one goal is to quantify patient benefit with directional microphones. When setting up this evaluation, a comparison of omnidirectional and directional amplification is completed. To verify the benefit of the directional processing, the hearing aid settings are identical except for the microphone mode. Many hearing aids offer an automatic (or adaptive) directional setting. These automatic features should be turned off and the directional system should be set in a static or "fixed" setting. The Destiny™, Zōn™ and S Series™ product lines provide an automatic system that estimates an environment's real-time signal-to-noise-ratio (SNR) and automatically switches between omni and directional microphone settings. However, due to the adaptive nature of the HINT test setup, it is advised that automatic directional systems be turned off to assure that the hearing aid is in the proper directional mode to quantify benefit.
Patients are seated one meter from each of eight speakers in a circular array during the HINT test. Speech is presented from the speaker located at 0 degrees azimuth relative to the patient's head. The remaining seven speakers present uncorrelated speech-shaped noise. See Figure 1 below for an illustration of this setup. This particular speaker array is used to simulate a more realistic and more difficult listening environment where background noise is diffusely distributed, compared to the standard HINT set-up where only a single speaker presents noise.
The clinician administering the test sits in an adjoining room and monitors the patient via a window and headset. The clinician controls the presentation of the test via a desktop computer. A GINA3G external sound card is utilized to process the computer generated stimuli and distribute it to the speaker array. The test administration is automated through the use of a Matlab script that automatically adjusts the presentation level of the sentences based on a correct or incorrect response from the listener. Figure 2 below shows a screenshot of the Matlab GUI that is seen on the computer screen during the test procedure. Signal presentation begins at a level of -10 dB SNR with the speech stimulus typically set at 55 dB SPL and the noise at 65 dB SPL. Threshold is defined as the SNR obtained after the presentation of 20 sentences. These data are then analyzed to evaluate listener performance as a function of SNR for each test condition.
EXAMPLE OF THE HINT IN A STARKEY CLINICAL TRIAL
To assess open-canal directional benefit, fourteen patients with mild to moderately-severe hearing loss completed a task of speech recognition in noise using the HINT. Each listener was fit bilaterally with Starkey receiver-in-the-canal hearing instruments to the prescriptive recommendations of NAL-NL1*, which is Starkey's proprietary modification of NAL-NL1 (Byrne et al., 2001). All of the fittings were completed with an open configuration, using the smallest non-custom earbud. Testing was conducted using the eight loudspeaker array described in the methodology section above. The following listening conditions were tested: 1) unaided 2) aided with omnidirectional processing and 3) aided with directional processing.
A one-way repeated measures analysis of variance revealed a significant main effect of listening condition, F(1,2) = 25.6, p = .001. Follow-up comparisons show a significant improvement in performance between the unaided and directional conditions (p<.001) and the omnizdirectional and directional conditions (p<.001). Figure 3 shows mean performance on the HINT as a function of unaided, omnidirectional, and directional listening conditions. The slight improvement in performance between the unaided and omnidirectional condition is not significant. The lack of significant improvement in the omnidirectional aided condition compared to the unaided condition is not surprising considering the relatively normal hearing in the low frequencies for most patients in the study and the lack of amplification with the open fit configuration in the low frequencies. Results show an average of 3.1 dB of improvement in performance threshold in the directional mode as compared to omnidirectional mode. The HINT results in this investigation provide evidence that open-canal directional processing can give listeners significant improvements to speech understanding in noise.
Bench, J., & Bamford, J. (Eds.) (1979). Speech-hearing tests and the spoken language of hearing-impaired children (Academic, London).
Byrne, D., Dillon, H., Ching, T., Katsch, R., & Keidser, G. (2001). NAL-NL1 procedure for fitting nonlinear hearing aids: characteristics and comparisons with other procedures. Journal of the American Academy of Audiology, 12, 37-51.
Hornsby, B.W.Y., & Ricketts, T.A. (2007). Effects of noise source configuration on directional benefit using symmetric and asymmetric directional hearing aid fittings. Ear & Hearing, 28(2), 177-186.
Kochkin, S. (2002). 10-year customer satisfaction trends in the US hearing instrument market. The Hearing Review, 9(10), 14-25, 46.
Nilsson, M., Soli, S.D., & Sullivan, J.A. (1994). Development of the hearing in noise test for the measurement of speech reception thresholds in quiet and noise. Journal of the Acoustical Society of America, 95(2), 1085-1099.
Nordrum, S., Erler, S., Garstecki, D., & Dhar, S. (2006). Comparison of performance on the hearing in noise test using directional microphones and digital noise reduction. American Journal of Audiology, 15, 81-91.
Ricketts, T.A., Lindley, G., & Henry, P. (2001). Impact of compression and hearing aid style on directional hearing aid benefit and performance. Ear & Hearing, 22(4), 348-361.