Speech-in-noise (SIN) testing provides a useful window into the status of a patient's auditory system. It can be used for clinical diagnosis and measurement of functional capacity of the hearing system, providing clinicians with highly valuable information while requiring minimal clinical time. However, SIN tests are infrequently used except where required for cochlear implants or auditory fitness-for-duty evaluations. As such, we explore the benefits of SIN tests, highlighting an inexpensive setup to complete SIN testing outside of a sound booth.
WHAT TESTS ARE AVAILABLE?
SIN tests are used to evaluate hearing ability in noise for both adults and children. For adults, sentence- and word-level tests are available. In English, the Quick Speech-In-Noise Test (QuickSIN; Etymotic Research) is the most commonly used test. It uses the IEEE sentence corpus divided into 12 groups of sentences that are played at a high signal-to-noise ratio (SNR) of +25 dB to a low SNR of 0 dB SNR. The stimulus is a female voice, and the noise is composed of a four-talker male and female babble. The QuickSIN takes one minute per list. Most clinicians use two or three sentences to obtain an SNR loss, which is the difference between a normal hearing population's performance and a patient's test performance (SNR to obtain 50 percent correct). The AzBio (Auditory Potential, LLC) sentence corpus—widely used especially in cochlear implant candidacy evaluations—uses conversational-style sentences presented at a fixed SNR and scored based on the percentage of correct keywords.1 The stimulus includes several male and female talkers, while the noise is a 10-talker babble. Other sentence-level tests in English include the Speech Perception in Noise Test-Revised (SPIN-R) and the Bamford-Kowell-Bench Speech-In-Noise (BKB-SIN) Test. At the word level, clinicians can use the Words-In-Noise (WIN) test. Previously, the Hearing In Noise Test (HINT) was also used, but it's no longer available for purchase.
Choosing a test for children can be challenging as the test must be appropriate for the child's language and cognitive levels. For older children (ages 5 and up), the BKB-SIN provides a similar SNR loss measure as the QuickSIN for adults. Like AzBio, BabyBio sentences can be used for children and presented at a fixed SNR. For younger children ages 3 to 6 years old, the Pediatric Speech Intelligibility (PSI) test, a closed-set task using picture pointing with a single competing talker, can be used.
A NOTE ON TEST LIMITATIONS
Clinicians should consider the patient's language and cognitive levels prior to administering a SIN test. For deaf and hard-of-hearing (DHH) children, language delays may limit the use of a specific SIN test. However, clinicians should not feel limited in trying SIN testing with children. In our clinic, the youngest child to complete the PSI was 2 years and 9 months old, and 3- to 4-year-old children are regularly tested in noise. Adults with developmental disabilities or acquired memory impairments may not be able to complete SIN testing at an age-appropriate level, and different tests may need to be employed. Additionally, patients whose first language is not the same as the language used in the test material may score more poorly than expected on the test. Some SIN tests are available in languages other than English, though clinicians should always ensure that their chosen test has been appropriately validated.
DIAGNOSTIC USE OF SIN TESTING
SIN tests can be used to understand a patient's auditory system status, providing a valuable look at the patient's complaints and the underlying physiology. The primary concern of many patients is difficulty hearing in noise, which pure tone audiometry in quiet, while useful for quantifying hearing sensitivity, cannot address. For these patients, a SIN test can provide useful information about their real-world complaint. The QuickSIN, for example, can be administered in as little as two minutes in a binaural condition. Ongoing research in our clinic and at Stanford University suggest that, in the future, audiologists may be able to replace word-recognition in quiet (WRS) with a QuickSIN test, which predicts WRS scores well. Additionally, we theorize that SIN tests may be used to identify a retrocochlear site of lesion when results are compared between ears; research is ongoing to evaluate this possibility.
Hidden hearing loss (HHL), a condition where patients report hearing difficulty in the presence of normal hearing thresholds (previously known as obscure auditory dysfunction), has been a recent hot topic in hearing science. For some patients with HHL, SIN tests will reveal a deficit not otherwise seen on an audiogram. Furthermore, most test batteries for central auditory processing disorders (APD) have used SIN tests for many years, calling this auditory figure-ground testing. In both cases, SIN test requires more auditory system resources and evaluates higher-order auditory system processes than those assessed using pure tone audiometry or speech perception in quiet.
FUNCTIONAL USE OF SIN TESTING
In addition to understanding a patient's auditory abilities, SIN tests have been widely recommended as a tool for both prediction of benefit from amplification and validation of benefit from amplification.2 For validating amplification benefit, SIN tests can be set up to compare aided and unaided hearing using either hearing aids or cochlear implants. Clinicians can choose between three types of SIN testing setups to best address a patient's concerns and evaluate hearing aid performance. On the left panel of Figure 1, speech and noise are spatially separated, which will take advantage of directional microphone benefits.
To predict benefit from hearing amplification, clinicians can interpret unaided SIN test results. For example, a patient with a high SNR loss on the QuickSIN may not show as much benefit from hearing aids as a patient with a low SNR loss. Patients with poor performance on SIN tests may benefit from technology that improves the real-world SNR, including directional microphones, remote microphone technology, or FM system technology. Patients with asymmetric hearing loss may also benefit from unilateral SIN testing. A patient with an acoustic neuroma, for example, may have a worse SNR loss in the ear with the neuroma. A patient with a high SNR loss in one ear may not find significant benefit in noise from unilateral amplification since the hearing aids may provide increased distortion rather than clarity of speech.
Evaluating the functional benefit of amplification using SIN tests can be particularly beneficial for patients with single-sided deafness (SSD) because the results of testing with demonstration devices can predict real-world outcomes.3 Snapp and Telischi's protocol for assessing patients with SSD requires patients to perform SIN tests with speech to the worse ear and noise to the better ear, with comparisons between aided and unaided conditions.4 The American Academy of Audiology has a similar guideline. Following this protocol allows the audiologist to predict both surgical and non-surgical outcomes for the patient with SSD and recommend the best hearing device for the patient. It can be used pre-intervention for hearing device selection and post-intervention for validation of hearing device benefit.
EVALUATING SIN OUTSIDE OF AUDIOMETRIC TESTS
A significant limitation of using SIN tests for hearing device validation is the accessibility of sound field systems in audiometric test suites. In busy audiology practices, clinicians may prioritize sound booth time for diagnostic audiometric evaluations or not have space in a sound booth to install a sound field system. In these cases, building a sound field system in a non-acoustically treated room may be a reasonable option to allow clinicians to use SIN tests more regularly. In practice, there is no significant limitation in using a quiet room instead of a sound booth for SIN testing since SIN tests are completed at a suprathreshold level. Here we present an inexpensive system that we use in our clinic's hearing aid fitting rooms.
As shown in Figure 2, this system comprises a single powered studio monitor speaker and a digital music player. A pair of speakers can also be used to split the stimulus and noise for utilizing directional microphones effectively. The digital music player must be able to play uncompressed files (such as .WAV files) or lossless compression files (such as .FLAC files). SIN test files can be ripped from a CD using free online software, then installed on the music player. Compressed file formats, including .MP3 or .AAC files as created by Apple products, should not be used for audiologic testing as they add a layer of compression that may result in test files that are not similar to the stimuli used in validation studies.
The clinician must then calibrate the system using a sound level meter to ensure that the stimulus is produced at a consistent level. We suggest the use of 60 dBA as a reasonable level for the validation of hearing device benefit. As shown in Figure 3, the patient should be seated at an appropriate distance to ensure the level is calibrated, and all volume at any levels should be fixed on the speakers and the music player. The system could also be used for tests of speech perception in quiet, such as the California Consonant Test, which may also help show the benefit from hearing devices.
The system shown in Figures 2 and 3 comprises a Samsung Clip Jam music player and a pair of Mackie CR3 monitor speakers; it costs less than $125, plus the cost of the QuickSIN test ($180 from Etymotic Research). The same system could be effectively put together with any high-quality speaker and digital music player. To evaluate the benefits of a directional microphone, the same system could be used with two speakers, one speaker in front and one behind the patient, and the split track version of the QuickSIN test. The benefits of validating hearing aids outweigh these reasonable costs.
Overall, SIN tests provide a significant benefit for audiology patients and should be regularly used for diagnostic and functional testing, as well as for validation of device benefit. Various SIN tests can be administered with little clinical time, proving to be valuable to both clinician and patient.