A Proven Strategy for Addressing Speech Understanding in Noise

This article was submitted to HR by Brian Taylor, AuD, manager of professional services at Unitron, Plymouth, Minn. Correspondence can be addressed to HR or .

A ccording to the latest MarkeTrak survey findings, an inability to understand speech in noise continues to be the primary complaint of hearing aid users. Many clinicians would agree that a failure to meet the needs of end users in background noise contributes to the reported 8% in-the-drawer rate and 14% total dissatisfaction rate for newer hearing aids. Unlike other manufacturers’ approaches to managing communication problems in background noise, Unitron has taken a unique approach.

A Decade of Similar Approaches

Traditional approaches to speech-in-noise problems include directional microphone strategies and various types of processor-based noise reduction algorithms (eg, digital noise reduction). Although directional microphones have been available for several decades, their efficaciousness was not validated until the mid-1990s when a series of studies indicated that, in ideal listening situations (ie, talker directly in front of the listener, noise directly behind, and conditions of low reverberation), the signal-to-noise ratio of the listening environment could be improved by 3 to 5 dB.

Since that time, several iterations of both directional microphone strategy and processor-based noise reduction algorithms have been brought to market with considerable promise and enthusiasm—only to encounter a lack of evidence supporting their effectiveness.

More recently, steering algorithms have been introduced that automatically switch the microphone from omni to directional or turn the digital noise reduction (DNR) from off to on. Because these automatic features are working independently of each other, their performance cannot be optimized across a variety of listening situations. For example, when the end user of a hearing aid with automatic switching walks into a noisy environment, the microphone is likely to switch from omni to directional, and the DNR feature switches from off to on. This is likely to result in sub-optimal performance in noise for most listeners.

FIGURE 1. A schematic of the laboratory test conditions used in the study. The improvement on the HINT and BKB-SIN from two sites for both the vented and open-canal condition are listed. Results (expressed as both dB and % improvement) compare hearing aids without smartFocus to the same product fitted with the smartFocus feature engaged.

A New Approach

While adaptive features have provided incremental improvements to overall ease of use for end users, all proven signal-to-noise ratio benefits have been associated with microphone strategy and FM technology.

Unlike the above traditional approaches to managing noise, smartFocus takes a unique approach by combining four features (microphone strategy, digital noise reduction, speech enhancement, and overall gain) as one algorithm. Tying the performance of these four features together allows them to work with maximum efficiency. For example, when an end user encounters background noise, smartFocus does the following:

1. Automatic directional microphones reduce off-target noise, which drastically improves the signal-to-noise ratio of the listening environment.
2. Therefore, the noise level of sound entering the hearing aid is lowered; thus, the Speech Enhancement processor is able to work more efficiently.
3. Digital Noise Reduction further reduces noise from the front of the listener, thus listening comfort is improved even more.

Evidence of Efficacy

A study completed at two independent and accredited universities proves that combining these four features in one algorithm results in remarkable synergetic effects. A total of 44 participants were involved at each site (University of Rochester and Louisiana Tech University) using the HINT and BKB-SIN. Half (22 participants) were fitted with traditionally vented instruments, while the other half were fitting with open-canal products. Testing for all subjects was completed in a sound booth that simulated real-world listening conditions.

As shown in Figure 1, using multi-talker babble from four azimuths and speech directly in front of the listener, speech recognition was compared at the default setting of a traditionally fitted premium instrument (no smartFocus) to settings in which smartFocus was activated. Results showed a 16% improvement in noise for traditionally vented instruments in the smartFocus “clarity” setting. Remarkably, open-canal devices—which many professionals believe do not offer the same improvements in noise as non-open products—were shown to have a 13.5% improvement in noise when smartFocus was activated.

Results of this study prove that, even at default settings, smartFocus significantly improves speech understanding in noise compared to similar products without it. Because smartFocus optimizes multiple adaptive features in relation to one another, end users can expect straight-out-of-the-box improvements with minimal manual adjustments using a remote control. SmartFocus is the first product in over a decade to significantly improve the signal-to-noise ratio (over traditional directional and FM) in an independent, replicable study.