Imagine yourself sitting in a fairly crowded restaurant. You are trying to listen to the person next to you on the right while there is a lot of interfering noise coming from the tables on the left. The speech signal from the talker would, no doubt, be stronger at the right ear than at the left ear due to head-shadow effect. Even then, with disturbing noise emanating from the left, the listener is likely to lean toward the speaker on the right, attempting to receive speech on the right side and ignore noise on the left. People with hearing loss would presumably have a greater need to segregate the desired speech signals from the constant stream of competing noise in such a situation.
To date, this complex need is yet to be fulfilled by advanced hearing instrument technology. In today’s bilateral fittings, noise reduction is applied to hearing instruments in both ears so long as noise is detected in both instruments. An undesirable consequence of this is that the gain for speech can be turned down on the side where the user has a speaker he or she wants to focus on. Even with wireless synchronization of noise management between ears, there is a possibility that the amount of noise attenuation on the noisy side may not be adequate to achieve optimal comfort and ease of listening.
Refining Binaural Processing
The solution to overcoming this limitation of current hearing instruments lies in a more intelligent use of binaural wireless technology.1 This technology—capable of sharing information about the listening environment between ears—can allow the hearing instrument user to turn the gain down on the “noisy” side, while at the same time turn up the gain on the “speech” side to maximize audibility of the speech signal.
Binaural wireless technology that uses spatial information shared between ears to amplify speech in one ear while attenuating noise on the opposite ear is known as Spatial Noise Management. Until the arrival of Oticon Agil, this technology has not been available in advanced hearing instruments.
Spatial Noise Management is designed to allow the user to better select and focus attention on the side with a more favorable signal-to-noise ratio (SNR) to maximize the audibility of speech on that side. In doing so, it has the potential to improve listening comfort and ease of listening in poor signal conditions.
Evidence supporting the benefit of this feature comes from two studies that are described in detail below. The first study, dealing with the preference for Spatial Noise Management, was completed internally in a laboratory at Oticon’s international headquarters in Denmark. The second study, investigating the performance of Spatial Noise Management in the laboratory and in the real world, was carried out at Oticon in Denmark and at the Hörzentrum, Oldenburg, Germany, and data from both sites were combined prior to analysis.
Preference for Spatial Noise Management
Study #1—Laboratory test. A total of 10 adult hearing instrument users (4 males and 6 females) ages 33 to 79 years (mean age 63.5 years) with mild to moderate hearing loss participated in this study. All participants were fit with Oticon Agil to the prescribed settings for testing. A two-alternative forced-choice method was used to document preference ratings of Spatial Noise Management. A touch screen was used to select between two settings in the hearing instruments: Spatial Noise Management ON versus Spatial Noise Management OFF. The study subjects were allowed to switch back and forth between the two settings as quickly and as many times as they wanted before making a decision about which setting they preferred. Hearing instruments were set to the omni-directional mode to avoid possible directional effects influencing test results.
|FIGURE 1. The two-speaker setup with speech presented from the right speaker at 62 dBSPL and white or pulsating noise presented from the left speaker at 64 dBSPL. Speakers were placed at ±50° azimuth.|
Testing was conducted in a sound treated booth. Two speakers placed at -50° and +50° azimuths were used; speech was presented from the right speaker and noise (white noise or pulsating noise) from the left speaker (Figure 1). The speech stimulus, a continuously looping 15-second clip of a female talker, was presented at 62 dBSPL. White or pulsating noise was presented at 64 dBSPL. The white noise had a flat spectrum up to 7 kHz and was stationary; the pulsating noise comprised 20 millisecond white noise pulses that were 220 milliseconds apart.
Participants were not informed what to listen for; they were simply told to listen to the speech and noise that were presented and to select the setting they preferred. Participants were also asked if they could hear a difference in the settings. If they could, they were asked to describe the difference and state why they preferred one setting over the other.
Participants overwhelmingly reported that they preferred the setting that was less noisy. For instance, when listening to speech and white noise, 7 out of 10 participants reported that their preferences were based on lower and/or more comfortable levels of noise.
|FIGURE 2. The preference for Oticon Agil with and without Spatial Noise Management for pulsating and white noise is indicated as a percent of the total trials.|
|FIGURE 3. The visual analog scale used in the listening effort measurement. Each participant was required to indicate with a vertical line how effortful he or she found it to be to understand the speech.|
|FIGURE 4. Median rating of listening effort for Oticon Agil (with Spatial Noise Management) and for the advanced non-wireless hearing instrument. Listening requires significantly less effort (p<0.05, Wilcoxon’s signed rank test) with Oticon Agil compared to the advanced non-wireless hearing instrument.|
|FIGURE 5. Mean rating of the real-world performance of hearing instruments in situations with speech on one side and noise on the opposite side (n=15). Oticon Agil was rated to be significantly better than the advanced non-wireless instrument (p<0.05).|
Results of this preference study showed that listeners can hear the difference between the two settings (Spatial Noise Management ON versus Spatial Noise Management OFF). They overwhelmingly preferred the Spatial Noise Management ON setting (Figure 2). Whether it was white or pulsating noise, the Spatial Noise Management ON setting was chosen at least 86% of the time.
This finding indicates that the intelligent use of spatial information between ears to emphasize speech on one side and turn down noise on the other in asymmetric listening conditions can possibly improve sound quality, listening comfort, and listening effort in noise for hearing instrument users. Of note, this benefit was experienced equally by test people using open and plus domes. (A total of 4 participants wore open domes, 1 wore one open and one plus, and 5 wore plus domes for testing.) This shows that the benefits of Spatial Noise Management are experienced with both open and closed fittings.
Study #2—Two-site laboratory and real-world field trial. Further evidence in support of Spatial Noise Management comes from a two-site study that was completed at Oticon’s international headquarters in Denmark and at the Hörzentrum, Oldenburg, Germany. In this study, a total of 39 experienced hearing instrument users (29 males and 10 females) with a flat to sloping hearing loss rated the listening effort when using Oticon Agil Pro and when using an advanced non-wireless hearing instrument.
Using a cross-over design, half the participants began the study with Oticon Agil and the other half with the advanced non-wireless hearing instrument. Participants used Oticon Agil for 2 weeks and the advanced non-wireless instrument for 2 weeks. After using both instruments for 4 weeks, participants’ listening effort to both instruments was evaluated in the laboratory.
The setup used to evaluate listening effort was identical to that used in the preference study (Figure 1), except for the speech stimuli used to evaluate listening effort. Five sentences from the Dantalle II test (Danish speech intelligibility in noise test) and the OLSA (German speech intelligibility in noise test) test were presented from the right speaker, and white noise from the left speaker. Participants were required to repeat each sentence, and then rate the difficulty of speech understanding with Oticon Agil and the advanced non-wireless hearing instrument following the presentation of all five sentences.
Participants were blinded to the hearing instruments, and the order of presentation of sentences was randomized between the two instruments. The signal-to-noise ratio for the listening effort measurement corresponded to 80% of the words being correctly recognized for each individual participant. Participants rated the instruments on a blank but a bound visual analog scale (Figure 3), which was later translated into one of the 7 descriptive categories of effort2: No effort, very little effort, little effort, moderate effort, considerable effort, much effort, and maximum effort.
A total of 15 of the 39 participants had reported they used Oticon Agil with Spatial Noise Management and the advanced non-wireless hearing instrument in at least one asymmetrical noise condition (speech on one side and noise on the other) in their daily environment during the course of the 2-week trial of the hearing instruments. They rated the real-world performance of each of the two instruments using a qualitative scale ranging from “very poor” to “very good” at the end of the 2-week trial period.
The results of the two-site study on listening effort and real-world hearing instrument performance in asymmetric noise conditions were in favor of Oticon Agil with Spatial Noise Management. The laboratory evaluation of listening effort showed the listening effort with Oticon Agil to be one category better than that with the advanced non-wireless hearing instrument (Figure 4). While participants spent “considerable effort” when using the advanced non-wireless hearing instrument to understand the five sentences, they expended only moderate effort when using Oticon Agil under the same conditions. This difference in the listening effort rating between the two instruments was statistically significant (p<0.05).
The real-world performance of Oticon Agil with Spatial Noise Management in asymmetric noise conditions (with speech on one side and noise on the other) was also rated to be significantly better (p<0.05) than that of the advanced non-wireless hearing instrument (see Figure 5). The average rating for Oticon Agil was “good,” whereas the rating for the advanced non-wireless hearing instrument was “acceptable.”
Laboratory and real-world data from the above studies clearly demonstrated the effectiveness of Spatial Noise Management in Oticon Agil in asymmetric noise conditions. Strong user preference and significantly better ratings of listening effort and real-world performance attest to this effectiveness.
- Neher T, Behrens T, Beck DL. Spatial hearing and understanding speech in complex environments. Hearing Review. 2008;15(12):22-25. Accessed July 15, 2010.
- Schulte M, Vormann M, et al. Listening effort scaling and preference rating for hearing instrument evaluation. Presented at: Workshop on Hearing Screening and New Technologies; Brussels; 2009. hearcom.eu/about/DisseminationandExploitation/Workshop.html. Accessed July 15, 2010.